Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
A 614 OULU 2013 UNIV ER S IT Y OF OULU P. O. BR[ 00 FI-90014 UNIVERSITY OF OULU FINLAND U N I V E R S I TAT I S S E R I E S SCIENTIAE RERUM NATURALIUM Professor Esa Hohtola HUMANIORA University Lecturer Santeri Palviainen TECHNICA ACTA WOLVERINE HABITAT SELECTION, DIET AND CONSERVATION GENETICS Postgraduate research fellow Sanna Taskila MEDICA Professor Olli Vuolteenaho SCIENTIAE RERUM SOCIALIUM University Lecturer Hannu Heikkinen SCRIPTA ACADEMICA Director Sinikka Eskelinen OECONOMICA Professor Jari Juga EDITOR IN CHIEF Professor Olli Vuolteenaho PUBLICATIONS EDITOR Publications Editor Kirsti Nurkkala ISBN 978-952-62-0226-6 (Paperback) ISBN 978-952-62-0227-3 (PDF) ISSN 0355-3191 (Print) ISSN 1796-220X (Online) UN NIIVVEERRSSIITTAT ATIISS O OU ULLU UEEN NSSIISS U Anni Koskela E D I T O R S Anni Koskela A B C D E F G O U L U E N S I S ACTA A C TA A 614 UNIVERSITY OF OULU GRADUATE SCHOOL; UNIVERSITY OF OULU, FACULTY OF SCIENCE, DEPARTMENT OF BIOLOGY A SCIENTIAE RERUM RERUM SCIENTIAE NATURALIUM NATURALIUM ACTA UNIVERSITATIS OULUENSIS A Scientiae Rerum Naturalium 614 ANNI KOSKELA WOLVERINE HABITAT SELECTION, DIET AND CONSERVATION GENETICS Academic dissertation to be presented with the assent of the Doctoral Training Committee of Technology and Natural Sciences of the University of Oulu for public defence in Keckmaninsali (Auditorium HU106), Linnanmaa, on 18 October 2013, at 12 noon U N I VE R S I T Y O F O U L U , O U L U 2 0 1 3 Copyright © 2013 Acta Univ. Oul. A 614, 2013 Supervised by Professor Jouni Aspi Docent Ilpo Kojola Reviewed by Professor Hannu Ylönen Doctor Katja Holmala Opponent Docent Janne Sundell ISBN 978-952-62-0226-6 (Paperback) ISBN 978-952-62-0227-3 (PDF) ISSN 0355-3191 (Printed) ISSN 1796-220X (Online) Cover Design Raimo Ahonen JUVENES PRINT TAMPERE 2013 Koskela, Anni, Wolverine habitat selection, diet and conservation genetics. University of Oulu Graduate School; University of Oulu, Faculty of Science, Department of Biology Acta Univ. Oul. A 614, 2013 University of Oulu, P.O. Box 8000, FI-90014 University of Oulu, Finland Abstract Rare and elusive species are difficult to study, because they are usually secretive, solitary, occur at low densities and have large home ranges. Wolverines (Gulo gulo) can both hunt and scavenge for food. In Fennoscandia, wolverines co-exist with either wild or semi-domesticated reindeer, which constitute their most important winter food. Approximately half of the 180–220 Finnish wolverines are found in northern Finland within the reindeer management area. However, the other half of the population is distributed in eastern and central Finland, and the ecology of wolverines especially in this area is poorly known. This research examined the habitat selection, diet and population genetics of wolverines in northern and eastern Finland. The results suggest that wolf presence is one of the most important variables influencing the habitat selection of wolverines. This finding supports the speculative idea that wolverines might benefit from being sympatric with wolves through increased scavenging opportunities. Furthermore, both the reproductive status of wolverines and the availability of different prey items were found to affect the wolverine diet. In northern Finland, semi-domesticated reindeer and mountain hare were the most frequently utilized prey species for breeding female wolverines. In eastern Finland, the most important food source for breeding females was moose carrion, whereas males and non-breeding females heavily utilized mountain hares. These results support the predictions of the optimal foraging theory, suggesting that wolverines opportunistically utilize the food source that is most energy-efficiently available. In areas with a low density of medium-sized ungulates, scavenging of wolf- and human-killed carrion plays an essential role in food acquisition by wolverines. According to the results of a population genetics investigation, two wolverine subpopulations exist in Finland: a northern and an eastern one. The overall genetic variability was found to be low, and signs of a recent population bottleneck were detected in both populations. It is likely that the wolverine populations in Finland would benefit from improved connectivity between them, but also with neighbouring populations in Scandinavia and north-eastern Russia. Keywords: carnivore guild, diet, habitat selection, population genetics, predation, scavenging, wolf, wolverine Koskela, Anni, Ahman habitaatinvalinta, ravinto ja populaatiogenetiikka. Oulun yliopiston tutkijakoulu; Oulun yliopisto, Luonnontieteellinen tiedekunta, Biologian laitos Acta Univ. Oul. A 614, 2013 Oulun yliopisto, PL 8000, 90014 Oulun yliopisto Tiivistelmä Ahma (Gulo gulo) tunnetaan sekä haaskansyöjänä että keskikokoisia hirvieläimiä ja pienriistaa saalistavana petona. Fennoskandian alueella peuran kesy tai villi muoto on ahman merkittävin saalislaji. Noin puolet Suomen 180–220 ahmasta elää Pohjois-Suomessa poronhoitoalueella, ja loput Itä- ja Keski-Suomessa. Poronhoitoalueen ulkopuolella elävien ahmojen ekologiaa on tutkittu erityisen vähän. Väitöstyössäni tarkastelin ahman habitaatinvalintaa, ruokavaliota ja populaatiogenetiikkaa pääasiassa Pohjois- ja Itä-Suomen alueilla. Tutkimukseni tulokset osoittivat, että suden läheisyys oli yksi tärkeimmistä ahman habitaatinvalintaan vaikuttavista tekijöistä. Tämä havainto tukee hypoteesia, jonka mukaan ahma saattaisi hyötyä susien läheisyydestä suuremman haaskatiheyden ansiosta. Todennäköisesti ahmat elävät mielellään samoilla seuduilla susien kanssa, mutta kaihtavat läheistä kanssakäymistä välttääkseen killansisäisen saalistuksen. Tutkimusteni perusteella sekä ahman lisääntymistila että alueen saaliseläinten saatavuus vaikuttivat ahman ruokavalioon. Poro ja metsäjänis olivat lisääntyvien ahmanaaraiden tärkein ravintokohde Pohjois-Suomessa. Itä-Suomessa merkittävin lisääntyvien naaraiden ravintokohde oli hirvi, jota ahmat hyödyntävät lähinnä haaskojen muodossa, mutta urokset ja ei-lisääntyvät naaraat sen sijaan saalistivat eniten metsäjänistä. Ahmat siis näyttävät hyödyntävän opportunistisesti sitä ravintokohdetta, joka kullakin alueella on energiatehokkainta saavuttaa. Susien ja ihmisten jälkeensä jättämät haaskat ovat merkittävä ravintolähde ahmoille alueilla, joilla keskikokoisten hirvieläinten tiheydet ovat alhaisia. Ahman populaatiogeneettinen tutkimus osoitti, että Suomen ahmat ovat geneettisesti jakautuneet kahteen alapopulaatioon, pohjoiseen ja itäiseen. Ahmakannan geneettinen monimuotoisuus oli pientä, ja molemmissa alapopulaatiossa oli nähtävissä merkkejä äskettäisestä pullonkaulailmiöstä. Populaatioiden välisen geenivirran määrän tulisi olla nykyistä korkeampi, jotta ahmakannan elinvoimaisuuden voisi katsoa olevan turvattu tulevaisuudessa. Asiasanat: ahma, haaskansyönti, habitaatinvalinta, petokilta, populaatiogenetiikka, ravinto, saalistus, susi To my parents 8 Acknowledgements The completion of my dissertation has been a long journey that was possible only with the support of several people. First of all, I would like to express my gratitude to my supervisors, Jouni Aspi and Ilpo Kojola, for their patient guidance, enthusiastic encouragement and useful critiques of this research. I also thank my co-authors, Rodrigo Esparza-Salas, Pekka Helle, Maija Joensuu, Salla Kaartinen and Seppo Rytkönen, for their valuable help. I am very grateful to Marja Hyvärinen, who collected a lot of the data used in this project and was a co-author in two papers. Furthermore, I would like to thank the reviewers of this study, Katja Holmala and Hannu Ylönen, for their valuable comments on my thesis and Roy Siddall for language revision. During the years, I have been privileged to work with a number of talented co-workers. I warmly thank Seppo Ronkainen for being a great help in the field and sharing his expertise on large carnivores with me. I am grateful to Samuli Heikkinen for helping me with telemetrical and geographical issues during the work. I also thank Alpo Komulainen, Eero Kortelainen and Ulla Poukkanen for offering the opportunity to trap wolverines on their privately-owned land. Furthermore, I thank the personnel of the University of Oulu, the Finnish Forest and Park Services and the Finnish Game and Fisheries Research Institute for their assistance with the data collection and analysis. Special thanks to those biologists who shared their expertise with me on species hair identification, especially the personnel of Taivalkoski Game and Fisheries Research Institute. Thanks to Maija Joensuu, who carried out the DNA analysis of the wolverine samples while working on her Master’s Thesis. I’m also very grateful to those local people who told me about their wolverine observations and gave me valuable tips on fresh wolverine tracks. I thank all my workmates in our department: Emma, Hilde, Laura, Mikko, Netta, Suvi, Tuomo, Vekku, and many others. I have been very lucky to have you around during this PhD project. Special thanks to Anna and Nelli, my Science Chicks fellows. Also, I am very grateful to my biology teachers in high school, Alpo Komulainen and Leena Korhonen, who taught me the basics of biology and encouraged me to continue my education at university. This thesis would not have been possible without financial support from several foundations. I am grateful to the Jenny and Antti Wihuri Foundation, Suomen Riistanhoito-Säätiö, Suomen Luonnonsuojelun Säätiö, University of Oulu Graduate School, and the Faculty of Science, University of Oulu, for 9 supporting my work. I also thank the Finnish Game and Fisheries Research Institute for providing the GPS collars used in this project. Above all, I warmly thank my family and friends for supporting my dreams and interests. I thank my sister and brother, Hanna and Antti, for encouraging me in every situation and sharing our weird sense of humour. I thank Heini, Liisa, Mirva, Niina, Paula, Päivi and Sanna for lifelong friendship. I’m fortunate to have spent some of my most cherished memories in your company. The good advice, support and friendship of my colleague, Laura Härkönen, has been invaluable on both an academic and a personal level, for which I am extremely grateful. I also thank Anne Keski-Jaskari, Inka Kynsilehto, Maija Jauhiainen, and many other friends for providing understanding company both during good and tough periods. Special thanks to my field assistant, curly-coated retriever Johka, who worked hard and with passion in the field. And, most of all, thanks to my parents. My father Markku has been a great help in the field work, and thanks to his knowledge and encouraging company I didn’t give up, even during the hardest moments. My mother Tuula has always supported my dreams and helped to solve the problems. Without you there is no way I could ever have gotten this far. Thanks for being there! A final tribute to those wolverines who “cooperated” in this project and whose tracks led me to fascinating sights, new experiences and unforgettable memories in nature: Eeki, Seita, Tahvo and many other intriguing wolverine individuals in the boreal forests of Finland. 10 List of original articles I Koskela A, Kaartinen S, Aspi J, Kojola I, Helle P & Rytkönen S (2013) Does grey wolf presence affect the habitat selection on wolverines? Annales Zoologici Fennici 50: 216–224. II Koskela A, Kojola I, Aspi J & Hyvärinen M (2013) The diet of breeding female wolverines (Gulo gulo) in two areas of Finland. Acta Theriologica 58: 199–204. III Koskela A, Kojola I, Aspi J & Hyvärinen M (2013) Effect of reproductive status on the diet composition of wolverines (Gulo gulo) in boreal forests of eastern Finland. Annales Zoologici Fennici 50: 100–106. IV Esparza-Salas R, Joensuu M, Koskela A, Ollila T & Aspi J Population genetic structure and genetic diversity of the wolverine (Gulo gulo) in Finland. (Manuscript) 11 12 Contents Abstract Tiivistelmä Acknowledgements 9 List of original articles 11 Contents 13 1 Introduction 15 1.1 Large carnivores in changing environment ............................................. 15 1.2 The role of large predators in ecosystems ............................................... 16 1.3 Intraguild interactions in large carnivore guilds...................................... 18 1.4 Study species ........................................................................................... 20 1.5 Genetic population structure of wolverines ............................................ 23 1.6 Aims of the study .................................................................................... 24 2 Material and methods 27 2.1 Study areas .............................................................................................. 27 2.2 Research methods ................................................................................... 28 2.2.1 Habitat selection (I) ...................................................................... 28 2.2.2 Foraging ecology (II, III) .............................................................. 30 2.2.3 Population genetics (IV) ............................................................... 31 3 Results and discussion 35 3.1 Effect of wolf presence on wolverine habitat selection (I)...................... 36 3.2 Effects of prey availability and reproductive status on the wolverine diet (II, III) ............................................................................. 38 3.3 Population genetics of wolverines (IV) ................................................... 42 4 Conclusions and recommendations for future research 47 References 49 Original articles 61 13 14 1 Introduction 1.1 Large carnivores in changing environment People and large carnivores have been in conflict throughout their common history (Woodroffe 2000, Linnell et al. 2001, Distefano 2005). Virtually all of the large mammalian carnivores have suffered range collapses during the last two centuries, mainly due to persecution, deforestation and habitat fragmentation (Mech 1995, Treves & Karanth 2003). The grey wolf (Canis lupus), brown bear (Ursus arctos) and Eurasian lynx (Lynx lynx) existed almost throughout Europe, but during the last century they disappeared from their southernmost distribution area (Boitani 2000, Breitenmoser et al. 2000, Swenson et al. 2000). The wolverine (Gulo gulo) also once occupied areas as far south as the southern parts of Scandinavia, Estonia, Lithuania, and north-east Poland, but has been restricted to a more reduced range (Landa et al. 2000). Small populations are vulnerable to extinction through random fluctuations and the loss of genetic diversity (Lande 1993, Foley 1994). Extinction rates are commonly greatest among large species (Jablonski & Raup 1995, MacFadden 2000). Therefore, small populations of large carnivores should be especially extinction-prone because of their trophic positions, which constrain them to ranging widely at low population densities. Colinvaux (1979) explained the rarity of large carnivores to be the natural result of Eltonian pyramids. Food webs can support only a few apex predators due to inefficiencies of converting energy produced by plants (Hutchinson 1959). However, human-induced mortality contributes more to the extinction of populations of large carnivores isolated in small reserves than do stochastic processes (Woodroffe & Ginsberg 1998). Nowadays, direct and indirect negative interactions with humans limit the abundance of carnivores. Linnell et al. (2001) found that populations of large carnivores increased after favourable legislation was introduced, despite an increase in human densities. Therefore, large carnivore persistence is probably more adequately explained by management policy than by human population density. Reduced hunting and increased ungulate populations might also explain the recolonization by large carnivores of their former ranges in North America and Europe (Massolo & Meriggi 1998, Oakleaf et al. 2006). Nowadays, the grey wolf, brown bear, Eurasian lynx and wolverine are globally considered as “least concern” species according to the criteria of the 15 International Union for Conservation of Nature (IUCN 2012). Globally, populations remain large, but locally there are many small, isolated populations. Large carnivores still face threats such as over-exploitation through legal and illegal hunting, small population risks such as inbreeding, and the fragmentation of habitats, with resulting areas becoming too small to ensure the long-term viability of populations. However, the European Mammal Assessment determined that the European wolverine is considered as “vulnerable”; thus, the least concern listing is driven by the estimation that some large populations remain in North Asia and North America. In Finland, brown bear and Eurasian lynx populations are considered as “vulnerable”, the grey wolf as “endangered” and the wolverine as “critically endangered” (for classification and criteria, see Rassi et al. 2010). 1.2 The role of large predators in ecosystems Large carnivores have the potential to place extraordinary pressures on their prey population, which in turn may lead to a variety of effects on the structure and function of ecosystems (Berger et al. 2001, Soulé et al. 2003, Estes et al. 2011). The structure of a community may be controlled by bottom-up or top-down forces. The bottom-up model postulates that it is mainly nutrients that limit the community by controlling plant numbers, which control herbivore numbers, in turn controlling predator numbers (Lindeman 1942). In contrast, according to the top-down model, consumers at higher trophic levels limit the abundance of lower trophic levels (Power 1992), and it is mainly predation that controls the community. Top carnivores are limiting consumers at lower trophic levels and create ripple effects throughout the food web. Such trophic cascades result from the top-down control of consumers over their prey (Paine 1980). Hairston et al. (1960) argued that since the world looks green, it is not overgrazed by herbivores (green world hypothesis). They concluded that carnivores and producers are controlled by density-dependent processes, but density-independent processes (e.g. predation) regulate the herbivores in the middle. Large carnivores typically have low population densities, large habitat area requirements and a position at the top of the food chain, and are thus often chosen as target species of conservation efforts (Ray 2005). Although the role of top predators in ecosystems is still not well understood, they are increasingly recognized as keystone species. Keystone species are single species that greatly affect communities but constitute only a small proportion of the community biomass (Paine 1966, 1969, Power et al. 1996). Most examples of keystone 16 predators are from either marine or freshwater aquatic ecosystems (Ray 2005), such as the sea otters (Enhydra lutris; Power et al. 1996), large predatory snails (Castilla & Duran 1985) or freshwater bass (Power et al. 1996). Terrestrial predator-prey interactions may be more difficult to study because they play out over much larger areas and over a much longer period of time (Ray 2005). The effects of terrestrial carnivores as keystone species can become evident when they are reintroduced to isolated ecosystems such as islands or national parks. The reintroduction of wolves into the Yellowstone National Park (Wyoming, USA) is a good example of top-down trophic cascades. During a period of 70 years without wolves in the park, notable changes in wildlife were the colonization of moose (Alces alces; Tyers 2003), elimination of beavers (Castor canadensis) and white-tailed deer (Odocoileus virginianus), reduction of aspen (Populus tremuloides), cottonwood (Populus deltoides) and willow (Salix sp.), and reduction of song birds (Hansen & Rotella 2002). Since their reintroduction within the park in 1995, wolves have mediated many effects throughout the food chain. The population size of moose has declined due to wolf predation, and coyote (Canis latrans) densities have diminished because of spatial avoidance and intraguild predation. At least 16 vertebrates have been shown to be affected in ecologically important ways by wolf predation (Ray 2005). Wolves have also facilitated the availability of carrion for at least 12 vertebrate scavengers, such as grizzly bears (Ursus arctor horribilis), golden eagles (Aquila chrysaetos) and ravens (Corvux corax). McLaren & Petersen (1994) also found that the density of wolves in Isle Royale National Park (Michigan, Canada) controls the abundance of moose, which in turn controls the abundance of the island’s balsam fir (Abes balsamea). In addition, the concept of umbrella species has been related to some large carnivores. Umbrella species are defined as species with large area requirements, which if given a sufficient protected area will bring many other species under protection (Noss 1990). An umbrella species needs large expanses of habitats and can therefore be used as a surrogate for many other biodiversity components that have similar but less extensive spatial requirements. Because of their large habitat area requirements, large carnivores are often seen as good umbrella species in conservation strategies. However, the ability of large carnivores to thrive in industrial forests means that the many species that are sensitive to modern forestry will not fall under the umbrella of areas managed for large carnivores (Linnell et al. 2000). 17 1.3 Intraguild interactions in large carnivore guilds An ecological guild is defined as a group of species that exploit the same resources in a similar way (Root 1967). On each continent, a guild of large carnivores has evolved, including members of the bear, dog, cat, hyena and mustelid families. However, predator pressure is often dominated by one species: the tiger (Panthera tigris) in Asia, the grey wolf in North America, and the lion (Panthera leo) in Africa. In Finland, the members of large carnivore guild are brown bear, grey wolf, Eurasian lynx and wolverine. Mesopredators (e.g. red fox, Vulpes vulpes; pine marten, Martes martes; golden eagle) also affect the overall predation pressure. Interactions between carnivores are usually categorized as competition (Peterson 1995, Creel & Creel 1996, Fedriani et al. 1999), predation (Palomares & Caro 1999, Linnell & Strand 2000) and mutualism or commensalism (DeVault et al. 2003, Selva 2004). Competition can occur and becomes important when two or more species strive for something that is not in adequate supply for both of them. The competition exclusion principle states that two species so similar that they compete for the same limiting resources cannot coexist in the same place (Gause 1934, Hardin 1960). Either the less competitive species will be driven to local extinction, or one of the species will evolve enough through natural selection to use a different set of resources. Interspecies competition has been recognized to be of two types, exploitation competition and interference competition. Exploitation competition is indirect and occurs through a common limiting resource. Competition for food is a form of exploitation competition in which carnivores interact with one another to access a shared prey species. Two large carnivore species in Finland, grey wolf and Eurasian lynx, are obligate predators, but they do not compete for the same prey species, as there are marked differences in their prey assortment (Pulliainen et al. 1995, Kojola et al. 2004). Two other species, the brown bear and wolverine, are seen as opportunistic predators and scavengers that can switch between food sources if one becomes rare. Interference competition occurs directly between individuals via aggressive behaviour and interspecific killing. Interspecific killing, or intraguild predation, is common among mammalian carnivores (Palomares & Caro 1999). Age, size and patterns of grouping play a significant role in the outcome of interactions between mammalian carnivores. Smaller species may kill cubs, young or subadult individuals of the larger species, but in general, larger species usually kill both 18 adult and juvenile individuals of the smaller species. For example, in North America, eight of the fourteen documented wolf–wolverine interactions resulted in death for the wolverines (Ballard et al. 2003). The reasons why wolves sometimes kill wolverines might include interspecific competition, hunger, territoriality or curiosity. Interference competition might result in avoidance behaviour and spatial segregation between carnivore species. May et al. (2008) found that wolverines differed in their habitat use compared to the three forest-dwelling carnivore species, as wolverines selected rugged terrain at higher elevations. This result fits well with the perception that susceptibility to fragmentation and human influence, and the avoidance of other carnivores affect wolverine habitat selection. In Finland, the elevational variation is low, and the availability of spatial refuges at higher elevations is therefore also low, especially in the eastern parts of the country. In addition, due to the scarcity of middle-sized ungulates in some areas, Finnish wolverines might greatly depend on moose carcasses in their diet, which may increase sympatry between wolverines and other large carnivores. Carnivores do not necessarily only have negative influences on each other, as intraguild interactions might even be beneficial for one (commensalism) or for both species (mutualism). Scavenging is a common form of foraging behaviour in the carnivore guild (DeVault et al. 2003, Selva 2004). Carcasses are considered as an ephemeral, rare and unpredictable resource in terms of their spatio-temporal availability (e.g. Heinrich 1988, Wilmers et al. 2003, Wilmers & Post 2006, Baglione & Canestrari 2009, DeVault et al. 2011). One foraging strategy that scavengers can employ to discover unpredictable food sources is to associate with other predators that provide food (Stahler et al. 2002). Carrion provided from the kills of predators and hunters may be much more predictable, especially for those scavenger species associating with large predators (Stahler et al. 2002, Wilmers et al. 2003, Selva et al. 2005). Other advantages of this type of a strategy include a reduced searching time, decreased energy expenditure and access to large, high quality food items (Brockmann & Barnard 1979, Heinrich 1988). On the other hand, predators can be a threat for scavengers (White et al. 2002, Ballard et al. 2003). The benefits gained from scavenging possibly compensate for the associated risk of being killed by a predator (Paquet 1992, van Dijk et al. 2008a, Gorini et al. 2012). The interactions between the members of Finnish large carnivore guild have been poorly studied and are probably not very strong because of the low population densities. Competition could occur between all species, as their diets 19 partly y overlap. The T largest sppecies, the brown b bear and grey wollf, might also o exhib bit occasional intraguild ppredation towards smaller guild g membeers. Scavenger speciies, the brow wn bear and w wolverine, dissplay commensal behaviou our by feeding g on caarrion killed by b more efficcient predatorrs (Fig. 1). Fig. 1. 1 Possible intteractions amo ong large carn nivores in Finland. 1.4 Study spe ecies With hin their holarrctic distribuution range, wolverines w oc ccupy a varieety of habitats rangiing across tun ndra, alpine aand taiga bio omes. The wo olverine is oft ften viewed as being g a high alpine dweller for which an nthropogenic factors, proobably due to o perseecution (Persson et al. 20009), are moree important in n the selectionn of the homee rangee than habitaat factors (Maay et al. 2006). Factors su uch as the deensity of prey y and landscapes where they could find a suitable denning d habbitat are also o impo ortant (Landa et al. 2000, M May et al. 20 010, May et al. 2012). A the end of the 16th cenntury, the disstribution ran At nge of wolver erines covered d almo ost the wholee Finland (P Pohja-Mykrä & Kurki 20 008). The poopulation sizee 20 started to decrease because of increased hunting pressure, and at the end of the 19th century the wolverine was common in northern Finland, but rare in other parts of the country (Ermala 2003). The estimated minimum population level, 50– 80 individuals, was probably reached before wolverines received protection in 1982. In 1979–1998, 16 wolverines were translocated from northern Finland to central parts of the country as an attempt to reduce reindeer (Rangifer tarandus tarandus) predation by wolverines and to expand the distribution area of wolverines to south (Pohja-Mykrä & Kurki 2008). Nowadays, approximately half of the estimated 180–220 Finnish wolverines are found in northern Finland within the reindeer management area, this subpopulation consisting of 80–90 individuals (Heikkinen & Kojola, in press). The other half of the population is distributed south of the reindeer management area, in eastern and central Finland. The subpopulation in eastern Finland consists of 50–60 individuals, whereas the estimated population size in central Finland is only 20–30 individuals (Heikkinen & Kojola, in press). Ungulate and carnivore densities differ considerably between the areas of the three subpopulations, which may be reflected in the foraging behaviour of wolverines. The eastern wolverine population has one unique feature: the population is located outside the reindeer management area in a forested habitat, where the density of medium-sized ungulates is low. This type of pattern can also be found in Sweden, where two isolated sub-populations have established in areas characterized by boreal forests with active forestry (Hedmark & Ellegren 2007). The minimum number of individuals in these populations was found to be 5 and 17, with one subpopulation showing increasing numbers and the other decreasing. The viability of these subpopulations in lowland forests is still unknown (Hedmark & Ellegren 2007). In Finland, about 80 wolverines are estimated to be living in boreal forests outside the reindeer management area, and even though the population is small, it has been stable during the last ten years (Heikkinen & Kojola, in press). This indicates that the wolverine is also capable of surviving in a habitat with a low number of medium-sized ungulates. The wolverine diet is varied and shows seasonal changes (Magoun 1987, Landa et al. 1997, Lofroth et al. 2007, van Dijk et al. 2008b, Dalerum et al. 2009). In North America and Scandinavia, the main food sources during the snow-free months are grouse species, mountain hares (Lepus timidus), small rodents and bird eggs. In winter, wolverines hunt reindeer and caribou and scavenge on ungulate carrion. The hunting success in winter is largely determined by the texture of the snow: if the snowpack is sufficiently hard to bear the weight of the 21 wolverine but not of the reindeer, the latter is an easy prey. Wolverines cache any surplus food and these reserves are important for their survival during periods when hunting is not possible or successful (Samelius et al. 2002, Wright & Ernst 2004). Species richness in most groups of organisms generally decreases towards the poles (Planka 1966, Currie & Paquin 1987, Kaufman 1995, Qian 1998), and a similar trend can also be found in Finland, with greater species diversity in southern Finland than in northern parts of the country. This may be reflected in differentiated foraging behaviour in wolverines, as there are more mesopredators and herbivores in eastern Finland than in the northernmost parts of the country. Wolverines have large home ranges, they occur at low densities and prey upon larger prey such as semi-domesticated reindeer. This predatory behaviour causes a conflict between wolverines and human interests. In northern Finland, the wolverine was responsible for most (51%) of the adult (<1 year old) reindeer predator-caused mortality between 2003–2008 (Norberg et al. 2010). Wolverine predation on semi-domesticated reindeer seems to be concentrated in the northernmost municipalities of northern Finland, and predation is heaviest during the late winter to early spring (Pohja-Mykrä & Kurki 2008). Reindeer depredation by large carnivores has been increasing during the last decade (Nordberg et al. 2010). On the contrary, in eastern Finland, predation by wolverines on domestic livestock is seldom reported. Between 2000–2006, the wolverine was responsible only on four reported cases of livestock losses outside the reindeer management area in Finland (Pohja-Mykrä & Kurki 2008). Wolverines have larger home ranges than other similar-sized carnivores. In North America and Europe, males and non-breeding females usually utilize the largest home ranges (200–1500 km2), whereas the area used by females with cubs ranges from 40 to 100 km2 (Whitman et al. 1986, Banci & Harestad 1990, Landa et al. 1998, Landa et al. 2000). Male wolverines roam within the ranges of several females. The wolverine is a polygamous species, and mating occurs from May to August (Wright & Rausch 1955, Rausch & Pearson 1972, Magoun & Valkenburg 1983, Mead et al. 1991). Wolverines exhibit delayed implantation (Wright & Rausch 1955), whereby the blastocyst does not immediately implant in the uterus but is maintained in a state of dormancy. Implantation usually occurs in November, with an effective gestation period of 30–50 days (Rausch & Pearson 1972). Most females give birth before late March in snow caves, and the average litter size is 2–3 cubs (Pulliainen 1968, Banci & Harestad 1988, Blomqvist 1995, Magoun & Copeland 1998, Persson et al. 2006). Young are weaned at 9–10 22 weeks and begin to travel with their mothers by late April. Mörner et al. (2005) found that the most common cause of death of wolverines in Sweden was traumatic injuries caused by other predators or wolverines, and human-caused mortality (killed by hunters in self-defence or suspected to be illegally shot). In North America, human-caused mortality accounted for 46% of deaths, and starvation was the most common natural mortality source (Krebs et al. 2004). The causes of wolverine mortality in Finland have not been studied. 1.5 Genetic population structure of wolverines The wolverine has small remnant populations in northern Europe, which are vulnerable due to persecution and conflict with human activities. It is unclear whether these populations are distinct from each other or whether they are connected to each other by migration. The isolation resulting from habitat fragmentation and the lack of connectivity can result in a decline in genetic variability and an increase in inbreeding. Reduced genetic variability can affect the reproductive potential and survival, which in turn can increase the extinction risk of a population (Frankham et al. 2002, Höglund 2009, Allendorf et al. 2013). The loss of genetic diversity can also be caused by a sudden or continuous reduction in population size, also known as a population bottleneck. Depending on the severity and duration of a bottleneck, rare alleles can be permanently lost in a population, even if it recovers from the bottleneck, and its evolutionary potential may consequently be reduced (Frankham et al. 2002, Höglund 2009). The low reproductive rate of the wolverine makes the species very sensitive to changes in survival rates. The deleterious effects of population reductions can be counteracted by increased migration. Immigrants from neighbouring populations can add new genetic variation to a population, preventing the extinction of small populations. Such a rescue effect has been observed, for example, in the Scandinavian wolf population (Vila et al. 2003). Wolverines exist in four countries in Europe (Norway, Sweden, Finland and Russia) and have two population strongholds: a “Scandinavian population”, which includes the areas of Norway, Sweden and far northern Finland, and a “Karelian population” including areas of eastern Finland and neighbouring parts of north-eastern Russia. It is likely that both populations are connected to each other, as wolverines are known to sporadically occur in areas between the two populations (Kaczensky et al. 2013). The combined population of wolverines for Norway, Sweden and Finland is estimated at approximately 1 250 individuals. 23 Census data from the Russian part of the Karelian population are not available (Kaczensky et al. 2013). The genetic population structure of wolverines in Scandinavia has previously been studied (Walker et al. 2001, Flagstad et al. 2004, Hedmark & Ellegren 2007), although the connection with the Karelian population has not been addressed. Extant Scandinavian populations appear to be subdivided and genetic variability was found to be lower than reported for other mustelid species (Walker et al. 2001). In North America, wolverine populations from southernmost regions, in which anthropogenic factors are strongest, were revealed to be more genetically structured than northern populations (Kyle & Strobeck 2001). Wolverines are highly mobile with the ability to disperse up to 300 km in one year (Gardner 1985, Magoun 1985, Gardner et al. 1986, Vangen et al. 2001, Cegelski et al. 2003). However, if the wolverine population in central Finland is isolated, it is likely that a population this small will face serious inbreeding problems. No information is available on genetic exchange between the central Finnish population and the Finnish-eastern Russian population. 1.6 Aims of the study The wolverine is one of Finland’s rarest mammals and least-studied large carnivores, with only few previous studies conducted in northern Finland (Pulliainen 1968, 1975). Even though the wolverine has been researched in North America and Scandinavia during recent years, knowledge of its ecology, population dynamics and genetics has remained limited. This thesis research focused on wolverine habitat selection (paper I), diet (papers II and III) and population genetics (paper IV). Wolverines, like other scavengers and opportunistic carnivores, can benefit from being sympatric with a more efficient predator (van Dijk et al. 2008b). One objective of this study was to investigate the relationship between wolverines and wolves. Understanding and predicting the interactions between the members of a large carnivore guild is crucial when trying to understand the ecology of species and manage their populations. In paper I, the aim was to investigate which factors affect the habitat selection of wolverines. Wolverine habitat selection is known to be negatively affected by human activity, including roads, infrastructure and backcountry recreation (Caroll et al. 2001, Rowland et al. 2003, May et al. 2006, Krebs et al. 2007). In this study, the possible effect of wolf presence on wolverine habitat selection was of 24 particular interest. Moreover, the effects of settlements and forest types on the habitat selection of wolverines were examined. Wolverines were predicted to prefer wolf presence and avoid settlements. In paper II, the diet of breeding female wolverines was investigated in two areas differing in the availability of prey. Wolverines were predicted to utilize the most available ungulate species, which are semi-domesticated reindeer in northern Finland and moose (carrion) in eastern Finland. Furthermore, a high density of medium-sized ungulates (semi-domesticated reindeer) in the northern Finland study area was expected to reduce the dietary niche breadth of wolverines. Paper III concentrates on possible sex differences in wolverine foraging ecology. The hypothesis was that breeding female wolverines, probably due to their smaller home ranges and higher energy demands (because of gestation and lactation), are forced to rely more on food caches and carrion than other wolverines. In paper IV, the aim was to investigate the genetic population structure, gene flow and genetic diversity of wolverines in remnant populations in Finland by examining microsatellite loci variation in DNA obtained by non-invasive sampling. One goal was to look for evidence of a population bottleneck in the Finnish wolverine populations and estimate the level of gene flow between the populations. Due to the small population size of wolverines in Finland, evidence of low genetic diversity and possible bottlenecks was predicted. Overall, this thesis research aimed to provide a better understanding of the ecology and genetics of Finnish wolverines. In this study, the unique ecology of wolverines living in eastern Finland outside the reindeer management area in forested habitat was of special interest. The results will provide a lot of practical information needed in securing the viability of small wolverine populations and preparing a management plan for Finnish wolverines. 25 26 2 Material and methods 2.1 Study areas The thesis is mainly based on data collected from two areas: within the reindeer management area in northern Finland and outside this area in eastern Finland (Fig. 2). The study area in northern Finland (Inari: 68°54’N, 27°01’E) belongs to the north-boreal coniferous forest zone (Ahti et al. 1968) and is composed of mountain plateaus with peaks of up to 600 m, consisting of bare rock and alpine tundra with heath and lichen vegetation. Lower elevation forests are dominated by mountain birch (Betula pubescens), Norway spruce (Picea abies) and Scots pine (Pinus sylvestris). The landscape in eastern Finland (Kajaani: 64°13’N, 27°43’E) belongs to the mid-boreal coniferous forest zone (Ahti et al. 1968) and is mainly composed of boreal forests, lakes and mires, the dominant tree species being the Scots pine and Norway spruce. Permanent snow usually appears in October/November and melts in May/June in northern Finland. In eastern Finland, the snow-cover period lasts from November/December to April. The elevation ranges from 100 to 600 m a.s.l. in northern Finland and from 100 to 300 m in eastern Finland. Temperatures vary from a minimum of -35 °C in winter to as high as 30 °C in summer. The annual precipitation ranges from 600 to 700 mm, with most of this falling during July and August. Both study areas are sparsely populated by humans; the mean density of humans is 0.4 people/km2 in the northern study area and 3.9 people/km2 in the eastern study area. Settlements and infrastructure are constrained near town centres, but many people maintain a holiday residence in the countryside. Extensive forest harvesting has occurred in both study areas, and the density of forestry roads is high. The study areas differ considerably in their ungulate and carnivore densities (Lavsund et al. 2003, Kojola et al. 2004, Wikman 2010, Nieminen et al. 2011). Free-ranging semi-domesticated reindeer dominate the ungulate population in the northern study area with a density of 2 animals/km2. The density of moose is 0.1 animals/km2. Other large carnivores are present at low densities (wolf and lynx: < 0.001 animals/km2, golden eagle: 72 breeding pairs). In the eastern study area, the density of wild forest reindeer (Rangifer tarandus fennicus) is 0.1–0.2 animals/km2 and that of moose 0.3–0.5 animals/km2. Densities of wolves and lynx are higher than in the northern study area (wolf 0.004 animals/km2, lynx 0.02 27 anim mals/km2). In both b study arreas, mountaiin hares, grou use and smalll rodents form m otherr possible foo od sources forr the wolverin ne. Fig. 2. 2 Study areas s. The wildlife e triangle datta (paper I) were w collected d from eastern n Finland, from the provinces of Kainuu, North hern Karelia and a Northern Savonia. Sca at ples for the dietary d studie es were collec cted from norrthern and eaastern Finland d samp (pape er II) and eastern Finland (p paper III). Mos st of the genotyped individu uals (paper IV V) were from northe ern and easttern Finland, but in addition two ind dividuals from m hwestern Rus ssia and one e individual from f the rein ntroduced weestern Finnish h north popu ulation were genotyped. g Th he samples frrom Russia were w from thee Pinega Stric ct Naturre Reserve, Arrkhangelsk reg gion (not show wn on the map p). 2.2 Research h methods 1 Habitat selection (I) 2.2.1 The habitat selecttion of wolveerines was sttudied by using wildlife trriangle snow wtrack k data and thee landscape ddata that were from the CORINE C Landd Cover 2000 0 datab base. Wildlife fe triangles arre equilaterall triangles with a perimetter of 12 km m, scattered throughout Finland. During the census, c the tra acks of mamm mals crossing g the trriangle line are a counted, aand a track in ndex and den nsity can be ccalculated. Alll geog graphic coord dinates of croossings of wo olverines and d wolves werre selected in n the wildlife w trian ngle data (oriiginal field forms) f from our study arrea in eastern n Finlaand (Fig. 2) in 2005–20008. As contrrol plots, an equal numbeer of random m crosssings were generated g from m existing wildlife w triangles in our sstudy area in n 28 order to determine whether wolverines seek areas closer to wolves rather than having a random proximity. Six landscape variables representing both infrastructural features and the main habitat types in Finland were selected to determine the factors behind wolverine habitat selection. The variables comprised six classes of land cover variables: (1) settlements, (2) mixed forests, (3) coniferous forests, (4) deciduous forests, (5) young forests and (6) mires. The seventh variable that was included in the analysis was the distance to the closest wolf crossing. Settlements were entered in the models as the distance of a given crossing or control plot from the nearest settlement. Other landscape variables were measured as the proportion of each habitat type around the crossing or control plot within a given radius. To determine the scale at which the landscape variables were most important in predicting wolverine occurrence, buffers of 75, 150, 300, 600, 1200, 2400 and 4800 m in radius around the wolverine and control plots were constructed. Generalized linear models were then fitted for each buffer radius using landscape variables as predictive variables. For later models, only the most significant buffer radius was used. Before any modelling, Spearman’s pairwise correlation analysis was conducted between the explanatory variables to identify possible multicollinearities. As there were no variables with a strong correlation (|r| > 0.6), they were all included in the models (Green 1979, Fielding & Haworth 1995, Sawyer et al. 2006, Klar et al. 2008). A set of 128 candidate models of the occurrence of wolverines and wolves were designed based on the two following general hypotheses: 1) wolverines seek areas in proximity to wolves (May et al. 2008, van Dijk et al. 2008a, van Dijk et al. 2008b); and 2) wolverines and wolves avoid the proximity of settlements where human presence is continuous (May et al. 2006, Krebs et al. 2007, Kaartinen et al. 2010). Logistic regression was used together with Akaike’s Information Criterion corrected for small sample sizes (AICc) to model the important habitats for wolverines and wolves in the study area. Selection of the best approximating models was based on the ΔAICc values, which were calculated as the differences between the current and minimum observed values of the model AICc (Burnham & Anderson 2002). The Akaike weights (wi) were also calculated for all models. The Akaike weight is a measure of the likelihood of the model for the given data (Johnson & Omland 2004), so that models with a larger wi better approximate the 29 data (see Burnham & Anderson 2002). The primary objective in model fitting was to assess the relative importance of the seven variables included. Akaike weights were summed across all the models in which individual variables occurred to obtain the relative importance of the variables. The larger the sum of the Akaike weights, the more important a variable is relative to the other variables considered. Model averaging was used to produce the best model, including aspects of a number of models (Greaves et al. 2006). In the set of best approximating models, parameter estimates were examined to assess the reliability of each variable as a predictor of the occurrence of a wolverine or wolf. Due to the uncertainty in model selection, the model-averaged slopes (β-values) of the variables and their standard errors were calculated using equations 4.1 and 4.9 in Burnham & Anderson (2002); models with ΔAICc < 4 were included. Statistical analyses were carried out using the free software R 2.12.1 (R Development Core Team 2010) with the package MuMIn (Barton Kamil 2010). 2.2.2 Foraging ecology (II, III) The diet of wolverines was investigated by analysing scats collected from wolverine denning sites (papers II and III) and by snow-tracking wolverines (paper III). In paper II, the dietary composition of breeding female wolverines in the reindeer management area (northern Finland) was compared with that of wolverines breeding outside this area (eastern Finland). Altogether, 421 scats collected from six active dens in northern Finland and four dens in eastern Finland during 2004–2006 were analysed (Fig. 2). In paper III, the dietary habits of wolverines in eastern Finland were examined by comparing the diet composition of breeding females with that of males and non-breeding females. Scats representing the winter diet of breeding females (n = 168) were collected from four wolverine den sites located in eastern Finland during 2004–2006. Scats representing the diet of males and non-breeding females (n = 64) were collected by tracking six wolverine individuals in eastern Finland during 2009–2011. Scats from den sites were collected after the denning period (from February/March to April/May), when the female wolverines and cubs had left their dens, and the scats thus represented the diet of the reproducing females during the denning period. Scats of males and non-breeding females were from the winter season, and by tracking these animals for long enough it was possible to determine that these scats did not belong to breeding females. 30 Collected scats were frozen and analysed at the University of Oulu. Each scat sample was washed in a 0.5-mm sieve under running water. The remains (hairs, feathers, bones, plant material, non-food items) were oven-dried at 80 °C for 24 hours. Hairs and feathers were identified to the species level using macroscopic and microscopic characteristics (Debrot et al. 1982, Teerink 1991, De Marinis & Asprea 2006) and a reference collection. The percentage occurrences of the prey species in scats were then calculated (van Dijk et al. 2007). In paper II, the occurrence of prey species in the diet was assessed using logistic mixed-effects models with the study area as a fixed covariate and the den as a random factor. In paper III, the reproductive status entered the model as a fixed covariate and the individual as a random factor. To measure how specialized or generalized wolverines are in their foraging behaviour, the diet breadth was calculated using the standardized Smith’s measure of niche breadth, FT, following equation 1: FT = Σ( ), (1) where FT = Smith’s measure of niche breadth pj = proportion of individuals found in resource state j (Σ pj = 1.0) aj = proportion of resource j is of the total resources (Σ aj = 1.0) (Smith 1982). The 95% confidence interval was calculated using equation 2: sin[arcsin(FT) ± 1.96/(2√ )] , (2) where FT = Smith’s measure of niche breadth N = total number of individuals studied. All statistical analyses were performed with SYSTAT version 13 (SYSTAT, Inc, Evanston, Illinois) and Microsoft Excel, version 2003. 2.2.3 Population genetics (IV) To study the population genetics of wolverines in Finland, 406 scat samples, ten muscle tissue samples and two oral mucosa swabs as sources of DNA were collected during 2004–2011, mainly in northern and eastern Finland (Fig. 2). The PSP Spin Stool DNA Plus Kit (Invitek) was used to collect and extract DNA from 31 the scat samples following the manufacturer’s instructions. DNA from the tissue samples was isolated using the DNeasy tissue kit (Qiagen) protocol for the purification of total DNA from animal tissues. To test whether the extracts had sufficient DNA quality and quantity, the extracts were amplified using the microsatellite marker Ggu 14 (Walker et al. 2001, Hedmark et al. 2004, Hedmark & Ellegren 2006). A total of twenty microsatellite loci were amplified from the samples that had a sufficient amount of DNA using a two-step multiplex polymerase chain reaction method. Primers DBY3Ggu and DBY7Ggu (Hedmark et al. 2004) designed from intron sequences of a Y-linked DBY gene were used as markers for sex identification. The ten microsatellite loci from multiplex groups 1 and 2 were used for the preliminary identification of individual wolverines using the Individual Identification analysis implemented in the program Cervus 3.0 (Kalinowski et al. 2007). The program uses a likelihood-based statistical method for parentage analysis and identity analysis for co-dominant genetic markers. The presence of null alleles, as well as of scoring errors in the microsatellite genotypes caused by stuttering, large allele dropout or typing mistakes was tested using the program MICRO-CHECKER 2.2.3 (van Oosterhout et al. 2004). The program identifies various genotyping and typographic errors by estimating the frequency of null alleles at a locus using a series of algorithms. The software STRUCTURE 2.3.4 (Pritchard et al. 2000, Falush et al. 2003, 2007, Hubisz et al. 2009) was used to infer the most likely number of populations (K) in the dataset. The program assigns individuals probabilistically to one of several assumed populations, characterised by their allele frequencies, using a Bayesian clustering method. The Bayesian spatial explicit model of population structure implemented in the program GENELAND 3.2.4 was used to detect possible areas of genetic discontinuity (Guillot et al. 2005). The program estimates the number of populations in a dataset as well as their spatial organisation based on allele frequency variation, using models that implement both geographic and genetic information. A graphical representation of the correspondence between allele scores and individual scores was produced with factorial correspondence analysis (FCA) as implemented in the program GENETIX 4.05 (Belkhir et al. 2004). The program DIYABC 1.0.4.46 (Cornuet et al. 2010) was used in order to infer the time at which population clusters had split from an ancestral population, as well as to infer the effective population sizes (Ne) of the assumed populations. The program makes inferences about population divergences, admixtures and changes in population size using approximate Bayesian computation. 32 An analysis of molecular variance (AMOVA) using Arlequin 3.5.1.3 (Excoffier et al. 2005, Excoffier & Lischer 2010) was performed to test the scenario of population subdivision and to calculate the contribution of the different population hierarchy levels to the observed level of genetic variation. The number of permutations used in the AMOVA analysis was 10 000. For this analysis, the scenario of two wolverine populations was tested: one in northern Finland and one in eastern Finland. Wright’s F-statistics were calculated for the analysis between populations, for the whole dataset and separately for each gender using the program FSTAT 2.9.3.2 (Goudet 2001) based on 10 000 permutations, and 95% confidence intervals were obtained by bootstrapping over loci. Expected heterozygosities and deviations from Hardy-Weinberg expectations were estimated using Genepop 1.4.10 (Raymond & Rousset 1995, Rousset 2008). Moreover, FSTAT was used to estimate allelic richness (AR) and inbreeding coefficient (Fis). The program estimates gene diversity per locus and sample using Nei’s (1987) unbiased estimator. Genetic diversity parameters and FIS were calculated for each locus in each population. The program BOTTLENECK 1.2.02 (Cornuet & Luikart 1996) was used to detect recent reductions in effective population size. The program detects an excess of heterozygotes, as would be expected under a scenario of recent population bottlenecks, by using different tests: a sign test, a standardized differences test and a Wilcoxon rank test. A spatial analysis of genetic diversity was performed with the program SPAGeDi 1.3 (Hardy & Vekemans 2002) to detect correlations between the kinship coefficient (Loiselle et al. 1995) and logarithmic geographical distance classes. The correlation was tested for the complete dataset, as well as individually for each gender. A difference between the slopes of the regression for each gender was tested using ANCOVA. The program GeneClass (Piry et al. 2004) was used to detect first-generation migrants. The ratio of the likelihood computed from the population from which the individual was sampled (L_home) over the highest likelihood value among all population samples, including the population from which the individual was sampled (L_home/L_max) (Paetkau et al. 2004), was chosen for likelihood computation. A Bayesian-based model from Rannala & Mountain (1997) was chosen for the criteria for computation. The probability that an individual is a resident was estimated with a resampling algorithm from Paetkau et al. (2004) by simulating 10 000 individuals. Long-term migration rates were estimated using 33 BayesAss 3 (Wilson & Rannala 2003). The program uses a Bayesian method relying on Markov chain Monte Carlo sampling to estimate immigration over the last several generations. Delta values of 0.15 were set for allele frequency, migration rate and inbreeding. The effective population size (Ne) was estimated using LDNe 1.31 (Waples & Do 2008). The program estimates contemporary effective population sizes using a biased correction for the Hill method (Hill 1981, Waples 2006) based on linkage disequilibrium data. Three estimates of Ne are given based on the lowest allele frequency used (0.01, 0.02 and 0.05). 34 3 Results and discussion According to the results, wolf presence was one of the most important variables influencing the habitat selection of wolverines (I). In eastern Finland, wolverines heavily utilized moose carrion provided by wolves and hunters, whereas semidomesticated reindeer were the most important food source in northern Finland (II). Breeding female wolverines opportunistically utilized carrion and hunted less on small prey than males and non-breeding females (III). According to the results of a population genetics investigation, two wolverine subpopulations exist in Finland, migration between the populations was quite low, and genetic diversity and effective population sizes were low (IV). A brief summary of the predictions, observations and conclusions is presented in Table 1. Table 1. Summary of predictions, results and conclusions concerning the four substudies. Study Prediction I Wolverines favour wolf presence. Observations Yes Conclusions Remote forest areas and the presence of wolves are important to wolverines in their habitat selection. Wolverines and wolves avoid settlements. II Yes Ungulates have an important role in the Wolverines opportunistically use the diet of breeding female wolverines. (ungulate) food source that is most energy-efficiently available. Yes Alternative, smaller prey and carcasses may replace ungulates in the wolverine Wolverines in northern Finland have a diet in areas with a low density of mid- smaller niche breadth than those living in eastern Finland. III size ungulates. No Breeding female wolverines use carrion Breeding females rely more on food opportunistically and hunt less on small caches and carrion than males and non-breeding females. Yes prey than males. The larger home ranges and greater mobility of male Breeding female wolverines have a wolverines probably enable them to smaller niche breadth than males and non-breeding females. IV more frequently come into contact with No genetic diversity. smaller prey than females. There are two wolverine subpopulations Wolverines are suffering from low Yes in Finland: northern and eastern ones. Signs of a recent population bottleneck were detected in both populations. Wolverine populations have experienced population bottlenecks. Yes Genetic variation was low. 35 3.1 Effect of wolf presence on wolverine habitat selection (I) The main result was that wolverines favoured wolf presence, coniferous forests, mixed forests and mires and avoided settlements, young forests and deciduous forests in their habitat selection. The model best explaining wolverine habitat selection included the presence of wolves, deciduous forests and distance from settlements as explanatory variables. These findings support the preliminary idea that wolverines might benefit from being sympatric with wolves through increased scavenging opportunities. Wolverines are known to utilize carcasses left by wolves (van Dijk et al. 2008a, 2008b) and lynx (Mattisson et al. 2011). On the other hand, van Dijk et al. (2008a) observed that wolverines do not use the tracks of other carnivores to locate carcasses. Accordingly, wolverines might occupy the same regions as wolves because of greater carrion availability, but not live in very close proximity to them in order to avoid intra-guild predation. Although the data did not distinguish the behavioural mechanisms behind the preferential association of wolverines with wolves, it is likely that both innate and learned behavioural responses towards wolves are involved. Wolves and wolverines generally selected the same type of habitat: they favoured coniferous and mixed forests and avoided settlements, young forests and deciduous forests. However, the relative importance of variables differed between the species. For wolverines, the most important variables predicting habitat selection were the presence of wolves, deciduous forests and distance from settlements (Table 2). The model best explaining wolf habitat selection consisted of settlements, coniferous forests and mixed forests. Table 2. The relative importance of variables in the wolverine habitat selection model. Variable Wolf Deciduous forests Settlements Mixed forests Coniferous forests Mires Young forests Relative variable importance 1.00 1.00 0.68 0.40 0.19 0.18 0.15 Our results indicate that both wolverines and wolves in eastern Finland avoid settlements. The selection of remote forest areas is consistent with previous studies on habitat selection and home range use by wolverines (May et al. 2006, Krebs et al. 2007) and wolves (Theuerkauf et al. 2003, Kaartinen et al. 2010). 36 Human activities have expanded and left less living space for wolverines, which appear to be very sensitive to human disturbance and also to habitat fragmentation (Carroll et al. 2001, Rowland et al. 2003). The disadvantageous effects of human activity on wolverines not only result from habitat loss, but also from the loss of prey species and other carnivore species that provide carrion (Landa et al. 2000). Poaching is one of the main causes of adult mortality for both species (Boitani 2000, Landa et al. 2000, Krebs et al. 2004, Mörner et al. 2005, Lovari et al. 2006, Persson et al. 2009), and the selected habitats in this study correspond to areas providing safety from human disturbance. The main distribution areas of wolverines in Finland are located in areas with a low human population density. In the future, wolverines might expand to new areas in the south, following the recolonizing wolves. Observations that people and predators can co-exist even at high human densities (Woodroffe 2000) suggest that the possible expansion of wolverines from north to south might be successful. Both study species selected forest types in a similar way: they favoured coniferous and mixed forests and avoided young and deciduous forests. According to LANDSAT satellite data, the majority (66%) of Finnish forests are mixed, with lower relative proportions of pure spruce (15%), pure pine (11%) and pure deciduous forests (8%) (Lindfors & Laurila 2000). Thus, the selected habitats might reflect the proportions of different forest types in Finland. Even though there were no variables with a strong correlation in the analysis, deciduous and young forests might be located closer to settlements than random, which might explain the avoidance of these forest types by both study species. The avoidance of young forests by wolves was, however, unexpected, because previous studies have demonstrated a tendency of wolves to select similar habitats to moose, their most important prey species (James et al. 2004). Global warming is expected to have profound effects on northern habitats. Concern over the decline of alpine habitats and associated species in climate change scenarios could have implications for wolverine conservation in the future. The wolverine is one of the 25 terrestrial mammal species in Canada whose northern distributions are bounded by the Arctic Ocean (Forsyth 1985). These species are not able to migrate further north because of the Arctic Ocean barrier (Kerr & Packer 1998). Wolverines, due to their widespread distribution and habitats in taiga forests, are not likely to become extinct or even rare with substantial climate change, but they are nevertheless likely to undergo significant losses of habitat, and may eventually require management intervention. 37 Snow is seen as an important part of the wolverine’s seasonal habitat requirement (Magoun & Copeland 1998, Aubry et al. 2007). Copeland et al. (2010) studied the hypothesis that the occurrence of wolverines is constrained by their obligate association with spring snow cover that persists throughout the denning period from February–May to April–March. They collected spatial information for 562 wolverine reproductive den sites in North America, Finland, Sweden and Norway, and found that 97.9% of the den sites occurred in areas that were snow-covered in at least 1 of 7 years. Scandinavian wolverines preferred den sites that were snow-covered in 6–7 of 7 years. All dens reported in Finland (n = 10) were determined to be snow dens, but 8 of them were located outside the satellite-based spring snow coverage area. According to the results, the wolverine’s southern range extends slightly beyond the spring snow coverage range, especially in areas with low topographic relief. It is possible that snow coverage is limiting the breeding success of wolverines in central Finland. 3.2 Effects of prey availability and reproductive status on the wolverine diet (II, III) Both the reproductive status of wolverines and prey availability in the area affected the wolverine diet. In northern Finland, the semi-domesticated reindeer and mountain hare were the most utilized prey species for breeding female wolverines. In eastern Finland, the most important food source of breeding females was moose, whereas males and non-breeding females heavily utilized mountain hares. A brief summary of the results concerning the diet of wolverines in Finland is presented in Table 3. 38 Table 3. Percentage occurrence of different prey species found in wolverine scats. The final rows give the Smith’s measure of niche breadth and confidence intervals (95%). Food source Northern Finland Eastern Finland Breeding females Breeding females Males and non-breeding Moose 3.7 55.2 22.0 Semi-domestic reindeer 65.9 0 0 Wild forest reindeer 0 0 3.3 Domestic pig 0 13.4. 0 Mountain hare 15.6 12.7 51.9 Grouse 7.4 11.0 14.8 Beaver 0 1.4 0 females Raccoon dog 0 1.4 3.1 Pine marten 0 1.3 0 Red fox 0 0 3.1 Rodents 6.3 2.3 0 Other 1.2 1.4 1.7 Number of scats 253 168 64 6 Number of wolverines 6 4 Niche breadth 0.640 0.763 0.680 CI% 0.29–0.89 0.37–0.98 0.40–0.88 Ungulates were an important part of the wolverine diet in all three groups (breeding females in northern Finland, breeding females in eastern Finland, and males and non-breeding females in eastern Finland). Ungulates (moose + semidomesticated reindeer) formed 70% of the wolverine diet in northern Finland. In eastern Finland, ungulate (moose + wild forest reindeer) remains occurred in 55% of the scats of breeding females. Males and non-breeding females in eastern Finland also utilized ungulates (25%). The occurrence of moose did not differ between the groups in eastern Finland. The proportion of moose in the diet of breeding females was statistically significantly higher in eastern Finland and that of reindeer was higher in northern Finland. The wolverine appears to use different foraging strategies depending on the area and prey availability. In northern Finland, wolverines utilize semidomesticated reindeer mainly by predation, as the densities of other carnivores are low in the area, and reindeer carrion is therefore only occasionally available. In eastern Finland, as moose are very seldom killed by wolverines, they scavenge from moose carrion killed by wolves and hunters. Prey caching is a common 39 foraging strategy of mustelids (Oksanen 1983, Oksanen et al. 1985, Wright & Ernst 2004), and wolverines may also benefit from food caches during periods of food scarcity (Vander Wall 1990, Pasitschniak-Arts & Larivière 1995). Persson (2005) suggested that carrion cached in late fall or early winter is an important determinant of the condition of female wolverines at the time of pregnancy and lactation and determines their reproductive success. As moose are harvested in autumn and early winter, it is evident that wolverines benefit from moose hunting by humans primarily before the breeding season, while wolf-killed moose is available for wolverines during all seasons. However, wolverines may also benefit from human-killed moose during the winter in the form of food caches. Although wild forest reindeers were present in small numbers in the eastern study area, there was no indication of wild forest reindeers being an important part of the wolverine diet. Only 3% of the diet of males and non-breeding females consisted of wild forest reindeers, and breeding females did not utilize them at all. When comparing semi-domesticated and wild forest reindeer, the latter might be a more elusive and challenging prey species for wolverines. However, there have been some observations of wolverines utilizing the carcasses of wild forest reindeer in eastern Finland, although the cause of mortality of reindeer in these cases is unknown (A. Koskela, unpublished data). The main wintering grounds of wild forests reindeer are located north of the eastern study area, and this might also explain the lack of the species in the diet of wolverines in our study. Nevertheless, as wolverines inhabit both of the main distribution areas of wild forest reindeer in Finland, in eastern and central Finland, it is reasonable to assume that wolverines also utilize wild forest reindeer at least in areas where the availability of other prey species or carrion is low. The presence of domestic pigs (Sus scrofa domesticus) in the diet of two breeding female wolverines in the eastern study area was most probably a consequence of wolverines visiting feeding sites at which pig carcasses had been placed to attract large carnivores for photographic purposes. Domestic pig constituted a notable proportion (52%) of the diet of one of the female wolverines. The distance between the den and the closest feeding site was 2.5 km. Persson (2005) found that the reproduction of wolverines is influenced by food availability in the winter, as reproduction was higher for food-supplemented females than for non-supplemented females. The feeding of large carnivores for photographic and recreational purposes is a common and increasing practice in eastern Finland, and this carrion supply might have a notable effect on the breeding success of wolverines. 40 The most frequently utilized food species for males and non-breeding females based on scat samples was the mountain hare, and mountain hare remains were also the most common carrion that was observed when tracking these wolverines. The diet of males and non-breeding females included more mountain hare than that of breeding females. These results support the predictions of the optimal foraging theory, suggesting that wolverines opportunistically utilize the food source that is most energy-efficiently available. Hunting ungulates or smaller prey such as mountain hare or grouse might incur greater energetic costs than scavenging, but searching for wolf- or human-killed carcasses takes more time. The larger home ranges and greater mobility of male wolverines probably enable them to more frequently come into contact with smaller prey than breeding females. In southern Norway, the main factor influencing the number of wolverine cubs surviving was the abundance of small rodents (Landa et al. 1997). Van Dijk et al. (2008b) also observed that small rodents formed 12% of the wolverine diet in forest habitats. However, our results demonstrated that small rodents were of negligible importance in the diet of breeding female wolverines (6% in northern Finland and 2% in eastern Finland). The proportion of small rodents in the wolverine diet was significantly higher in northern Finland than in the eastern study area. Small rodent remains were not found from the scats of males or nonbreeding females. Voles in northern Europe have been shown to exhibit cyclic population dynamics. The length of the vole population cycle is reported to be 3– 5 years, with generally synchronous fluctuations in different vole species (Hansson and Henttonen 1985, 1988). During 2004–2006, when the scats of breeding females were collected for this study, vole densities were very low in eastern Finland (Huitu et al. 2009). In northern Finland, vole population densities were very high in 2005, but in 2004 and 2006 the densities of small rodents were low (Huitu et al. 2009). Vole population cycles might therefore explain the small proportion of rodents in the diet of breeding female wolverines in our study. The niche breadth was highest in the diet of breeding females in eastern Finland (0.76) and lowest in the diet of breeding females in northern Finland (0.64) (Table 2). However, because the confidence intervals overlapped, the niche breadth did not differ significantly between the groups. Nevertheless, the diet of wolverines in eastern Finland contained more prey species than that of wolverines in northern Finland, indicating that wolverines in northern Finland are more specialized in hunting semi-domesticated reindeers and are not opportunistic scavengers to the same extent as wolverines in eastern Finland. 41 In Finland, the wolf population has been declining for the last five years. Although the wolverine population size in Finland has remained stable during recent years, decreasing wolf numbers might have a negative effect in the long term. At the same time, other sources that provide carrion to wolverines might compensate the effects of the declining wolf population. Wolverines might also utilize brown bear kills outside the hibernation season. In Sweden, wolverines have been found to benefit from being sympatric with lynx through increased scavenging opportunities (Mattisson et al. 2011, Rauset 2013). The lynx population has been strongly increasing in Finland during the last decade. Moose and other carcasses also become available for wolverines through hunter harvesting, ecotourism, vehicle collisions and natural death. The survival rate of adult females is generally one of the most important parameters for the population growth of long-lived animals (Nelson & Peek 1982). These results demonstrated the great importance of carrion, especially to breeding female wolverines during the breeding season. Therefore, if the goal is to facilitate the colonization of wolverines from northern Finland to eastern and southern parts of the country, the highest priority should be given to ensuring the carrion supply by maintaining the wolf population size at a viable level in the wolverine distribution area. 3.3 Population genetics of wolverines (IV) Altogether, DNA was extracted from 406 scat samples, from which 211 DNA isolates amplified well enough to be visible in agarose gels, giving a success rate of 52%. After the individual analysis of all available genotypes, we were able to identify a total of 67 individuals. We were able to correctly assign the gender to 61 individuals. The likelihood values obtained with STRUCTURE were highest for four populations, although the delta K values suggest the most likely number of populations is two. The individuals largely split into two groups, with one population including the northern Finland individuals and one western Finland individual (hereafter “northern population”) and another population that included samples from eastern Finland (hereafter “eastern population”). The two individuals from Pinega, Russia were grouped together with the northern population. The Bayesian spatially explicit analysis of Geneland supported a scenario of two populations, with a rather clear boundary between them. The results for the analysis of molecular variance suggest that most of the observed 42 variation could be explained by differences within the populations (96%), and to a lesser extent by differences between the northern and eastern populations (4%). The fixation index (FST) value of 0.042 was highly significant, which indicates a significant differentiation between the two populations. It is unlikely that such a population structure is an artefact of strong isolation by distance with non-sampled individuals (Frantz et al. 2006), as has been found in other carnivores (Aspi et al. 2006, Kopatz et al. 2012), since this study included samples covering much of the latitudinal distribution in the area. The observed population structure in such a relatively small area could be explained by post-glacial colonization, which is characteristic of a number of species in Fennoscandia (Pamilo & Savolainen 1999), whereby some genotypes originated from glacial refuges in central or eastern Europe and arrived through eastern Finland, whereas others originated in western or central Europe and arrived via past land bridges connecting the Jutland and Scandinavian peninsulas. The sample from Western Finland was also assigned to the northern population and was not genetically differentiated enough to cluster on its own, suggesting that the western Finnish wolverines have not differentiated enough from the northern population (from which they originate due to translocations) to form a genetically distinct population. However, these results are not reliable due to the small sample size, although they offer an interesting topic for future research. The F-statistics (Weir & Cockerham 1984) across all loci were estimated with the program FSTAT. The analysis gave an FIT value of 0.025, a pairwise FST value of 0.046 and an FIS value of -0.022. The F-statistics were estimated separately for both sexes using the program FSTAT. The program gave males an FIS value of 0.088 and an FST value of 0.066. For females, the FIS value was 0.023 and the FST value 0.048. The p-value for the FST test was 0.714, suggesting that there is no male-biased dispersal according to these data. The fixation index FST between the Finnish populations was the same magnitude as the global FST reported between wolverine populations in Scandinavia (Walker et al. 2001), and within the confidence intervals of those estimated more recently between a main population and two recently founded populations in Scandinavia (Hedmark & Ellegren 2007). The FST was more similar to that recorded between fragmented populations in North America than to that of more continuously connected populations within Alaska (Kyle & Strobeck 2001). The observed and expected heterozygosities, as well as allelic richness were found to be similar in both populations, although the observed heterozygosity was 43 slightly higher. The mean inbreeding coefficients were slightly negative and not significantly different from zero in either population. The expected heterozygosities appeared to be relatively low and similar to those reported in Scandinavian populations (Flagstad et al. 2004, Hedmark & Ellegren 2007), although generally lower than in larger North American populations (Kyle & Strobeck 2001). Heterozygosities were lower than those found in other large carnivores in the same areas in Finland (Aspi et al. 2006, Kopatz et al. 2012). The observed differentiation between populations could be due to the effects of drift in small populations, which can be particularly intense in populations that have experienced severe population bottlenecks (Barton & Charlesworth 1984). Evidence of recent population bottlenecks was found in both populations, with the assumption of 70% stepwise mutation in all three tests. Under the assumption of 90% stepwise mutation, we found significant values for both populations using the Wilcoxon test, but not using the sign test or the standardized differences test, although some of the results were somewhat close to significance. The program GeneClass detected one immigrant (female) from the eastern to the northern population and two migrants (one female and one individuals of unknown sex) from the northern to the eastern population. The position of these first-generation migrants was evident in the factorial correspondence analysis. Long-term migration estimated using the Bayesian approach detected global nonmigration rates of 0.833% and migration rates of 0.166%. Between-population migration estimates were estimated at around 0.30 in either direction, and were not significantly different from each other. Global and between-population migration rates were significantly different from zero, as indicated by the 95% confidence intervals, which implies a low to moderate level of migration. There was some evidence of gene flow between the two populations, as indicated by the migration results and the presence of individuals in either population that appeared to have an ancestry from another population. This indicates that the populations are not completely isolated from one another, although the gene flow has not been large enough to homogenize the allele frequencies in both populations. The effective population size was 14.2–16.8 in the eastern population and 14.4–17 in the northern population depending on the lowest allele frequency used. The values were low and similar in both populations. The estimated effective population sizes were considerably lower than the census population size, and lower than the Ne value of 50, which has been suggested as the minimum size to 44 avoid serious loss of genetic variation in the short term (Soulé & Kohm 1989, Allendorf & Ryman 2002). The results of this study demonstrate that the extant wolverine populations in Finland suffer from relatively low genetic diversity, potentially resulting from a recent population bottleneck. It is likely that the wolverine populations in Finland would benefit from improved connectivity between them, but also with neighbouring populations in Scandinavia and north-eastern Russia. There appears to be an increasing trend locally in the European wolverine populations, but the low effective population sizes and the evidence of population bottlenecks suggest that these populations have a larger extinction risk and are vulnerable to population reductions caused by humans or by stochastic events. 45 46 4 Conclusions and recommendations for future research The results presented in this thesis suggest that a multitude of factors influence the foraging behaviour of wolverines. Both the reproductive status of wolverines and the prey availability in the area, as well as the amount of carrion, affected the wolverine diet. The foraging ecology of wolverines markedly differed between northern and eastern Finland. Wolverines preyed heavily on semi-domesticated reindeer in northern parts of the country, whereas in eastern Finland wolverines tended to be mainly scavengers. Outside the reindeer management area, breeding females utilized moose carrion, whereas males and non-breeding females hunted small prey species. There is considerable interest in understanding the role of scavengers on the community level. Scavengers might cause predators to kill more, by stealing food remains, and strengthen the top-down effect on vegetation through the control of ungulate populations. Furthermore, wolverines should be seen not only as scavengers utilizing carrion left by other carnivores, but also as predators offering carcasses to smaller carnivores. In the rugged habitats of arctic, sub-arctic and boreal forest zones, where the number of large carnivore species is low, the impacts of wolverines on the ecosystem require further study. In addition to reproduction status and prey availability, the presence of another large carnivore, the grey wolf, affected the wolverine diet. The present results revealed that wolf presence was one of the most important variables influencing the habitat selection of wolverines. It appeared that wolverines occupied the same regions as wolves because of greater carrion availability, and that the benefits of scavenging for wolverines were worth the risk of aggression from wolves. The importance of scavenging in the foraging behaviour of wolverines indicates that maintaining a wolverine population in the presence of wolves is ecologically feasible. Keeping the wolf and wolverine populations viable is a management challenge that will test wildlife managers in Finland. The presence of the grey wolf was an important factor influencing the wolverine diet, but it is reasonable to assume that wolverines also benefit other carnivore species, such as the brown bear, Eurasian lynx, red fox and raptors. Therefore, the nature of wolverine interactions with other large carnivores requires further study. The overall genetic variability was low in both wolverine populations in Finland according to the heterozygosity values when compared with other wolverine, Mustelidae and carnivore populations. Estimates of the effective 47 population size were also low in both populations and below the suggested threshold value for populations to avoid the loss of genetic variation in the short term. One of the future management goals should be to increase the effective population sizes and genetic variability of the Finnish wolverine populations to ensure their viability. Furthermore, the level of connectivity with other extant population clusters in northern Europe needs to be resolved in order to propose management decisions that are relevant to the whole region. 48 References Ahti T, Hämet-Ahti L & Jalas J (1968) Vegetation zones and their sections in northwestern Europe. Ann Bot Fenn 5: 169–211. Allendorf F, Luikart G & Aitken S (2013) Conservation and the genetics of populations. Oxford UK, John Wiley & Sons. Allendorf F & Ryman N (2002) The role of genetics in population viability analysis. In: Bessinger S & McCollough D (eds.) Population Viability Analysis. Chicago, The University of Chicago Press: 50–85. Aspi J, Roininen E, Ruokonen M, Kojola I & Vilá C (2006) Genetic diversity, population structure, effective population size and demographic history of the Finnish wolf population. Mol Ecol 15: 1561–1576. Aubry KB, McKelvey KS & Copeland JP (2007) Distribution and broadscale habitat relations of the wolverine in the contiguous United States. J Wildlife manage 71(7): 2147–2158. Baglione V & Canestrari D (2009) Kleptoparasitism and temporal segregation of sympatric corvids foraging in a refuse dump. Auk 126: 566–578. Ballard WB, Carbyn LN & Smith DW (2003) Wolf interactions with non-prey. In: Mech D & Boitani L (eds) Wolves: behaviour, ecology and conservation. Chicago, The University of Chicago Press: 259–271. Banci V & Harestad AS (1988) Reproduction and natality of wolverine (Gulo gulo) in Yukon. Ann Zool Fenn 25: 265–270. Banci V & Harestad AS (1990) Home range and habitat use of wolverines Gulo gulo in Yukon, Canada. Ecography 13(3): 195–200. Barton N & Charlesworth B (1984) Genetic revolutions, founder effects and speciation. Annu Rev Ecol Evol Syst 15: 133–164. Belkhir K, Borsa P, Chikhi L, Raufaste N & Bonhomme F (2004) Genetix 4.05, logiciel sous WindowsTM pour la génétique des populations. Montpellier France, Laboratoire Génome, Populations, Interactions, CNRS UMR 5000, Université de Montpellier II. Berger J, Stacey PB, Bellis L & Johnson MP (2001) A mammalian predator-prey imbalance: grizzly bear and wolf extinction affect avian neotropical migrants. Ecol Appl 11: 947–960. Blomqvist L (1995) Reproductive parameters of wolverine (Gulo gulo) in captivity. Ann Zool Fenn 32: 441–444. Boitani L (2000) Action plan for the conservation of wolves (Canis lupus) in Europe. Strasbourg, Council of Europe Press. Breitenmoser U, Breitenmoser-Würsten C, Okarma H, Kaphegyi T, Kaphegyi-Wallmann U & Müller UM (2000) Action plan for the conservation of the Eurasian lynx in Europe (Lynx lynx). Strasbourg, Council of Europe Press. Brockmann HJ & Barnard CJ (1979) Kleptoparatism in birds. Anim Behav 27: 487–514. Burnham KP & Anderson DR (2002) Model selection and multimodel inference. A practical information-theoretic approach. 2nd ed. New York, Springer. 49 Carroll C, Noss R & Paquet P (2001) Carnivores as focal species for conservation planning in the Rocky Mountain Region. Ecol Appl 11: 961–980. Castilla JC & Duran LR (1985) Human exclusion from the rocky intertidal zone of central Chile: the effects on Concholepas concholepas (Gastropoda). Oikos 45: 391–399. Cegelski CC, Waits LP & Anderson J (2003) Assessing population structure and gene flow in Montana wolverines (Gulo gulo) using assignment-based approaches. Mol Ecol 12: 2907–2918. Colinvaux P (1979) Why big fierce animals are rare: an ecologist’s perspective. Princeton USA, Princeton University Press. Copeland JP, McKelvey KS, Aubry KB, Landa A, Persson J, Inman RM, Krebs J, Lofroth E, Golden H, Squires JR, Magoun A, Schwartz MK, Wilmot J, Copeland CL, Yates RE, Kojola I & May R (2010) The bioclimatic envelope of the wolverine (Gulo gulo): do climatic constrains limit its geographic distribution? Can J Zool 88: 233–246. Cornuet JM & Luikart G (1996) Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144: 2001–2014. Cornuet JM, Ravigne V & Estoup A (2010) Inference on population history and model checking using DNA sequence and microsatellite data with the software DIYABC (v1.0). BMC Bioinformatics 11: 401. Creel S & Creel NM (1996) Limitations of African wild dog by competition with larger carnivores. Conserv Biol 10: 526–538. Currie DJ & Paquin V (1987) Large-scale biogeographical patterns of species richness of trees. Nature 329: 326–327. Dalerum F, Kunkel K, Angerbjörn A & Shults BS (2009) Diet of wolverines (Gulo gulo) in the western Brooks Range, Alaska. Polar Res 28: 246–253. Debrot S, Fivaz G, Mermod C & Weber JM (1982) Atlas des poils mammifères d’Europe. Institut de Zoologie. Switzerland, Université de Neuchâtel. (In French) De Marinis AM & Asprea A (2006) Hair identification key of wild and domestic ungulates from southern Europe. Wildlife Biol 12: 305–320. DeVault TL, Olson ZH, Beasley JC & Rhodes OE (2011) Mesopredators dominate competition for carrion in an agricultural landscape. Basic Appl Ecol 12: 268–274. DeVault TL, Rhodes OE & Shivik JA (2003) Scavenging by vertebrates: behavioral, ecological, and evolutionary perspectives on an important energy transfer pathway in terrestrial ecosystems. Oikos 102: 225–234. Distefano E (2005) Human-wildlife conflict worldwide: a collection of case studies, analysis of management strategies and good practices. SARD Initiative Report. Rome, FAO. Ermala (2003) A survey of large predators in Finland during the 19th–20th century. Acta Zool Lit 13: 15–20. 50 Estes JA, Terborgh J, Brashares JS, Power ME, Berger J, Bond WJ, Carpenter SR, Essington TE, Holt RD, Jackson JB, Marquis RJ, Oksanen L, Oksanen T, Paine RT, Pikitch EK, Ripple WJ, Sandin SA, Scheffer M, Schoener TW, Shurin JB, Sinclair AR, Soulé ME, Virtanen R & Wardle DA (2011) Trophic downgrading of planet Earth. Science 333: 301–306. Excoffier L, Laval G & Schneider S (2005) Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evol Bioinform Online 1: 47–50. Excoffier L & Lischer HEL (2010) Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Res 10: 564–567. Falush D, Stephens M & Pritchard JK (2003) Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics 164 : 1567–1587. Falush D, Stephens M & Pritchard JK (2007) Inference of population structure using multilocus genotype data: dominant markers and null alleles. Mol Ecol Notes 7: 574– 578. Fedriani JM, Palomares F & Delibes M (1999) Niche relations among three sympatric Mediterranean carnivores. Oecologia 121: 138–148. Fielding AH & Haworth PF (1995) Testing the generality of bird-habitat models. Conserv Biol 9: 1466–1481. Flagstad O, Hedmark E, Landa A, Broseth H, Persson J, Andersen R, Segerstrom P & Ellegren H (2004) Colonization history and noninvasive monitoring of a reestablished wolverine population. Conserv Biol 18 : 676–688. Foley P (1994) Predicting extinction times from environmental stochasticity and carrying capacity. Conserv Biol 8: 124–137. Forsyth A (1985) Mammals of the Canadian wild. Camden East Ontario, Camden East Publishing. Frankham R, Ballou JD & Briscoe DA (2002) Introduction to conservation genetics. Cambridge UK, Cambridge University Press. Frantz AC, Pourtois JT, Heuertz M, Schley L, Flamand MC, Krier A, Bertouille S, Chaumont F & Burke T (2006) Genetic structure and assignment tests demonstrate illegal translocation of red deer (Cervus elaphus) into a continuous population. Mol Ecol 15: 3191–3203. Gardner CL (1985) The ecology of wolverines in southcentral Alaska. MSc thesis. University of Alaska. Gardner CL, Ballard WB & Jessup RH (1986) Long distance movement by an adult wolverine. J Mammal 67: 603. Gause GF (1934) The struggle for existence. Baltimore MD, Williams & Wilkins. Gorini L, Linnell JDC, May R, Panzacchi M, Boitani L, Odden M & Nilsen EB (2012) Habitat heterogeneity and mammalian predator–prey interactions. Mammal Rev 42: 55–77. 51 Goudet J (2001) FSTAT, a program to estimate and test gene diversities and fixation indices. Version 2.9.3. URI: http://www.unil.ch/izea/softwares/fstat.html. Cited 2013/05/16. Greaves RK, Sanderson RA & Rushton SP (2006) Predicting species occurrence using information-theoretic approaches and significance testing: An example of dormouse distribution in Cumbria, UK. Biol Conserv 130: 239–250. Green RH (1979) Sampling design and statistical methods for environmental biologists. New York, John Wiley & Sons. Guillot G, Mortier F & Estoup A (2005) GENELAND: a computer package for landscape genetics. Mol Ecol Notes 5: 712–715. Hairston NG, Smith FE & Slobodkin LB (1960) Community structure, population control and competition. Am Nat 94: 421–425. Hansen AJ & Rotella JJ (2002) Biophysical factors, land use, and species viability in and around nature reserves. Conserv Biol 16(4): 1112–1122. Hansson L & Henttonen H (1985) Gradients in density variations of small rodents: the importance of latitude and snow cover. Oecologia 67: 394–402. Hansson L & Henttonen H (1988) Rodent dynamics as community processes. Trends Ecol Evol 8: 195–200. Hardin G (1960) The competitive exclusion principle. Science 131: 1292–1297. Hardy OJ & Vekemans X (2002) SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Mol Ecol Notes 2: 618– 620. Hedmark E & Ellegren H (2006) A test of the multiplex pre-amplification approach in microsatellite genotyping of wolverine faecal DNA. Conserv Genet 7: 289–293. Hedmark E & Ellegren H (2007) DNA-based monitoring of two newly founded Scandinavian wolverine populations. Conserv Genet 8: 843–852. Hedmark E, Flagstad Ø, Segerström P, Persson J, Landa A & Ellegren H (2004) DNAbased individual and sex identification from wolverine (Gulo gulo) faeces and urine. Conserv Genet 5: 405–410. Heikkinen S & Kojola I (in press) Suurpetojen lukumäärä ja lisääntyminen vuonna 2012. In: Wikman M (ed.) Riistakannat 2012. Riistaseurantojen tulokset. Helsinki, Riista- ja kalatalouden tutkimuslaitos. (In Finnish) Heinrich B (1988) Winter foraging at carcasses by three sympatric corvids with emphasis on recruitment by the raven, Corvus corax. Behav Ecol Sociobiol 23: 141–156. Hill WG (1981) Estimation of effective population size from data on linkage disequilibrium. Genet Res 38: 209–216. Hubisz MJ, Falush D, Stephens M & Pritchard JK (2009) Inferring weak population structure with the assistance of sample group information. Mol Ecol Res 9: 1322– 1332. Huitu O, Kiljunen N, Korpimäki E, Koskela E, Mappes T, Pietiäinen H, Pöysä H & Henttonen H (2009) Density-dependent vole damage in silviculture and associated economic losses at a nationwide scale. Forest Ecol Manag 258: 1219–1224. 52 Hutchinson GE (1959) Homage to Santa Rosalia: or why are there so many kinds of animals. Am Nat 93: 145–159. Höglund J (2009) Evolutionary conservation genetics. USA, Oxford University Press. IUCN (2012) The IUCN Red list of threatened species. Version 2012.2. URI: http://www.iucnredlist.org. Cited 2013/05/16. Jablonski D & Raup DM (1995) Selectivity of end-Cretaceus marine bivalve extinctions. Science 268: 389–391. James ARC, Boutin S, Hebert DM & Rippin AB (2004) Spatial separation of caribou from moose and its relation to predation by wolves. J Wildlife Manage 68: 799–809. Johnson JB & Omland KS (2004) Model selection in ecology and evolution. Trends Ecol Evol 19: 101–108. Kaartinen S, Luoto M & Kojola I (2010) Selection of den sites by wolves in boreal forests in Finland. J Zool 281: 99–104. Kaczensky P, Chapron G, von Arx M, Huber D, Andrén H & Linnell J (2013) Status, management and distribution of large carnivores – bear, lynx, wolf & wolverine – in Europe. Rome, Istituto di Ecologia Applicata. Kalinowski ST, Taper ML & Marshall TC (2007) Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol Ecol 16: 1099–1006. Kaufman DM (1995) Diversity of New World mammals: universality of the latitudinal gradients of species and bauplans. J Mamm 76: 322–334. Klar N, Fernández N, Kramer-Schadt S, Herrman M, Trinzen M, Büttner I & Niemitz C (2008) Habitat selection models for European wildcat conservation. Biol Conserv 141: 308–319. Kojola I, Huitu O, Toppinen K, Heikura K, Heikkinen S & Ronkainen S (2004) Predation on European wild forest reindeer (Rangifer tarandus) by wolves (Canis lupus) in Finland. J Zool 263: 229–235. Kopatz A, Eiken HG, Hagen SB, Ruokonen M, Esparza-Salaz R, Schregel J, Kojola I, Smith ME, Wartiainen I, Aspholm PE, Wikan S, Rykov AM, Makarova O, Polikarpova N, Tirronen KF, Danilov PI & Aspi J (2012) Connectivity and population subdivision at the fringe of a large brown bear (Ursus arctos) population in North Western Europe. Conserv Genet 13: 681–692. Kerr J & Packer L (1998) The impact of climate change on mammal diversity in Canada. Environ Monit Assess 49: 263–270. Krebs J, Lofroth E, Copeland J, Banci V, Cooley D, Golden H, Magoun A, Mulders R & Shults B (2004) Synthesis of survival rates and causes of mortality in North American wolverines. J Wildlife Manage 68: 493–502. Krebs J, Lofroth E & Parfitt I (2007) Multiscale habitat use by wolverines in British Columbia, Canada. J Wildlife Manage 71(7): 2180–2192. Kyle CJ & Strobeck C (2001) Genetic structure of North American wolverine (Gulo gulo) populations. Mol Ecol 10: 337–347. Landa A, Lindén, M & Kojola I (2000) Action plan for the conservation of wolverines (Gulo gulo) in Europe. Strasbourg, Council of Europe Press. 53 Landa A, Strand O, Linnell JD & Skogland T (1998) Home range sizes and altitude selection for arctic foxes and wolverines in an alpine environment. Can J Zool 76: 448–457. Landa A, Strand O, Swenson JE & Skogland T (1997) Wolverine and their prey in southern Norway. Can J Zool 75: 1292–1299. Lande R (1993) Risks of population extinction from demographic and environmental stochasticity and random catastrophes. Am Nat 142: 911–927. Lavsund S, Nygrén T & Solberg EJ (2003) Status of moose populations and challenges to moose management in Fennoscandia. Alces 39: 109–130. Lindeman RL (1942) The trophic dynamic aspect of ecology. Ecology 23: 399–418. Lindfors V & Laurila T (2000) Biogenic volatile organic compound (VOC) emissions from forests in Finland. Boreal Environ Res 5: 95–113. Linnell JDC & Strand O (2000) Interference interactions, co-existence and conservation of mammalian carnivores. Divers Distrib 6: 169–176. Linnell JDC, Swenson JE & Andersen R (2000) Conservation of biodiversity in Scandinavian boreal forests: large carnivores as flagships, umbrellas, indicators or keystones? Biodivers Conserv 9: 857–868. Linnell JDC, Swenson JE & Andersen R (2001) Predators and people: conservation of large carnivores is possible at high human densities if management policy is favourable. Anim Conserv 4: 345–349. Lofroth EC, Krebs JA, Harrower WL & Lewis D (2007) Food habits of wolverine Gulo gulo in montane ecosystems of British Columbia, Canada. Wildl Biol 13: 31–37. Loiselle BA, Sork VL, Nason J & Graham C (1995) Spatial genetic structure of a tropical understory shrub, Psychotria officinalis (Rubiaceae). Am J Bot 82: 1420–1425. Lovari S, Sforzi A, Scala C & Fico R (2006) Mortality parameters of the wolf in Italy: does the wolf keep himself from the door? J Zool 272: 117–124. MacFadden BJ (2000) Cenozoic mammalian herbivores from the Americas: reconstructing ancient diets and terrestrial communities. Annu Rev Ecol Syst 31: 33–59. Magoun AJ (1985) Population characteristics, ecology and management of wolverines in northwestern Alaska. PhD thesis. University of Alaska. Magoun AJ (1987) Summer and winter diets of wolverines, Gulo gulo, in arctic Alaska. Can Field Nat 191: 392–397. Magoun AJ & Copeland JP (1998) Characteristics of wolverine reproductive den sites. J Wildlife Manage 62: 1313–1320. Magoun AJ & Valkenburg P (1983) Breeding behaviour of free-ranging wolverines (Gulo gulo). Acta Zool Fenn 174: 175–177. Massolo A & Meriggi A (1998) Factors affecting habitat occupancy by wolves in northern Apennines (northern Italy): a model of habitat suitability. Ecography 21: 97–107. Mattisson J, Persson J, Andrén H & Segerström P (2011) Temporal and spatial interactions between an obligate predator, the Eurasian lynx (Lynx lynx), and a facultative scavenger, wolverine (Gulo gulo). Can J Zool 89: 79–89. 54 May R, Gorini L, van Dijk J, Brøseth H, Linnell JCD & Landa A (2012) Habitat characteristics associated with wolverine den sites in Norwegian multiple-use landscapes. J Zool 287: 195–204. May R, Landa A, van Dijk J, Linnell J & Andersen R (2006) Impact of infrastructure on habitat selection on wolverines Gulo gulo. Wildl Biol 12: 285–295. May R, van Dijk J, Landa A, Andersen R & Andersen R (2010) Spatio-temporal ranging behaviour and its relevance to foraging strategies in wide-ranging wolverines. Ecol Model 221: 936–943. May R, van Dijk J, Wabakken P, Swenson J, Linnell J, Zimmermann B, Odden J, Pedersen H, Andersen R & Landa A (2008) Habitat differentiation within the large-carnivore community of Norway’s multiple-use landscapes. J Appl Ecol 45: 1382–1391. McLaren BE & Petersen RO (1994) Wolves, moose, and tree-rings on Isle Royale. Science 266: 1555–1558. Mead RA, Rector M, Starypan G, Neirinckx S, Jones M & DonCarlos MN (1991) Reproductive biology of captive wolverines. J Mamm 72: 807–814. Mech DL (1995) The challenge and opportunity of recovering wolf populations. Conserv Biol 9: 1–9. Mörner T, Eriksson H, Bröjer C, Nilsson K, Uhlhorn H, Ågren E, Hård af Segerstad CH, Jansson DS & Gavier-Widén D (2005) Diseases and mortality in free-ranging brown bear (Ursus arctos), gray wolf (Canis lupus), and wolverine (Gulo gulo) in Sweden. J Wildlife Dis 41: 298–303. Nei M (1987) Molecular evolutionary genetics. New York, Columbia University Press. Nelson LJ & Peek JM (1982) Effect of survival and fecundity on rate of increase of elk. J Wildl Manage 46: 535–540. Nieminen M, Norberg H & Maijala V (2011) Mortality and survival of semi-domesticated reindeer (Rangifer tarandus tarandus L.) calves in northern Finland. Rangifer 31: 71– 84. Norberg H, Kojola I & Härkönen S (2010) Petovahinkojen tunnistamisopas. Äänekoski, Metsästäjäin Keskusjärjestö, Keski-Suomen Painotuote Oy. (In Finnish) Noss RF (1990) From endangered species to biodiversity. In: Kohm K (ed.) Balancing on the brink: A retrospective on the endangered species act. Washington, Island Press: 227–246. Oakleaf JK, Murray DL, Oakleaf JR, Bangs EE, Mack CM, Smith DW, Fontaine JA, Jimenez MD, Meier TJ & Niemeyer CC (2006) Habitat selection by recolonizing wolves in the northern Rocky Mountains of the United States. J Wildlife Manage 70: 554–563. Oksanen T (1983) Prey caching in the hunting strategy of small mustelids. Acta Zool Fenn 174: 197–199. Oksanen T, Oksanen L & Fretwell SD (1985) Surplus killing in the hunting strategy of small predators. Am Nat 126: 328–346. Paetkau D, Slade R, Burden M & Estoup A (2004) Genetic assignment methods for the direct, real-time estimation of migration rate: a simulation-based exploration of accuracy and power. Mol Ecol 13: 55–65. 55 Paine RT (1966) Food web complexity and species diversity. Am Nat 100: 65–75. Paine RT (1969) A note on trophic complexity and community stability. Am Nat 103: 91– 93. Paine RT (1980) Food webs, linkage interaction strength, and community infrastructure. J Anim Ecol 49: 667–685. Palomares F & Caro TM (1999) Interspesific killing among mammalian carnivores. Am Nat 153: 492–508. Pamilo P & Savolainen O (1999) Post-glacial colonization, drift, local selection and conservation value of populations: a northern perspective. Hereditas 130: 229–238. Paquet PC (1992) Prey use strategies of sympatric wolves and coyotes in Riding Mountain National Park, Manitoba. J Mammal 72: 337–343. Pasitschniak-Arts M & Larivière S (1995). Gulo gulo. Mamm Species 499: 1–10. Persson J (2005) Female wolverine (Gulo gulo) reproduction: reproductive costs and winter food availability. Can J Zool 83: 1453–1459. Persson J, Ericsson G & Segerström P (2009) Human caused mortality in the endangered Scandinavian wolverine population. Biol Conserv 142: 325–331. Persson J, Landa A, Andersen R & Segerström P (2006) Reproductive characteristics of female wolverines (Gulo gulo) in Scandinavia. J Mammal 87: 75–79. Peterson RO (1995) Wolves as interspesific competitors in canid ecology. In: Carbyn LN, Fritts SH & Seip DR (eds) Ecology and conservation of wolves in a changing world. Edmonton Canada, Canadian Circumpolar Institute: 315–323. Piry S, Alapetite A, Cornuet JM, Paetkau D, Baudouin L & Estoup A (2004) GeneClass2: A software for genetic assignment and first-generation migrant detection. J Hered 95: 536–539. Planka ER (1966) Latitudinal gradients in species diversity: a review of concepts. Am Nat 100: 33–46. Pohja-Mykrä M & Kurki S (2008) Asialistalla ahma. Helsingin yliopisto, Ruraliainstituutti: 13. (In Finnish with English summary) Power ME (1992) Top-down and bottom-up forces in food webs: do plants have primacy? Ecology 73: 733–746. Power ME, Tilman D, Estes JA, Menge BA, Bond WJ, Mills LS, Daily G, Castilla JC, Lubchenko J & Paine RT (1996) Challenges in the quest for keystones. BioScience 46: 609–620. Pritchard JK, Stephens M & Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155: 945–959. Pulliainen E (1968) Breeding biology of the wolverine (Gulo gulo) in Finland. Ann Zool Fenn 5: 338–344. Pulliainen E, Lindgren E & Tunkkari P (1995) Influence of food availability and reproductive status on the diet and body condition of the European lynx in Finland. Acta Theriol 40: 181–196. Pulliainen E & Ovaskainen P (1975) Territory marking by a wolverine (Gulo gulo) in northeastern Lapland. Ann Zool Fenn 12: 268–270. 56 Qian H (1998) Large-scale biogeographic patterns of vascular plant richness in North America: an analysis at the generic level. J Biogeogr 25: 829–836. Rannala B & Mountain JL (1997) Detecting immigration by using multilocus genotypes. Proc Nat Acad Sci USA 94: 9197–9201. Rassi P, Hyvärinen E, Juslén A & Mannerkoski I (2010) The 2010 Red List of Finnish Species. Helsinki, Ympäristöministeriö & Suomen ympäristökeskus. Rausch RA & Pearson AM (1972) Notes on the wolverine in Alaska and the Yukon territory. J Wildlife Manage 36: 249–268. Rauset GR (2013) Life and death in wolverines. PhD Thesis. Swedish University of Agricultural Sciences. Ray JC (eds.) (2005) Large carnivores and the conservation of biodiversity. Covelo California, Island Press. Raymond M & Rousset F (1995) GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. J Hered 86: 248–249. Root RB (1967) The niche exploitation pattern of the blue-grey gnatcatcher. Ecol Monogr 37: 317–350. Rousset F (2008) Genepop'007: a complete reimplementation of the Genepop software for Windows and Linux. Mol Ecol Resour 8: 103–106. Rowland M, Wisdom M, Johnson D, Wales B, Copeland J & Edelman F (2003) Evaluation of landscape models for wolverines in the interior northwest, United States of America. J Mammal 84: 92–105. Samelius G, Alisauskas RT, Larivière S, Bergman C, Hendrickson CJ, Phipps K & Wood C (2002) Foraging behaviours of wolverines at a large arctic goose colony. Arctic 55: 148–150. Sawyer H, Nielson RM, Lindzey F & McDonald LL (2006) Winter habitat selection of mule deer before and during development of a natural gas field. J Wildlife Manage 70: 396–403. Selva N (2004) The role of scavenging in the predator community of Bialowieza Primeval Forest. PhD thesis. Polish Academy of Sciences and University of Sevilla. Selva N, Jedrzejewska B, Jedrzejewski W & Wajrak A (2005) Factors affecting carcass use by a guild of scavengers in European temperate woodland. Can J Zool 83: 1590– 1601. Smith EP (1982) Niche breadth, resource availability, and inference. Ecology 63: 1675– 1681. Soulé ME, Estes JA, Berger J & del Rio CM (2003) Ecological effectiveness: conservation goals for interactive species. Conserv Biol 17: 1238–1250. Soulé ME & Kohm KA (1989) Research priorities for conservation biology. Covelo California, Island Press. Stahler D, Heinrich B & Smith D (2002) Common ravens, Corvus corax, preferentially associate with grey wolves, Canis lupus, as a foraging strategy in winter. Anim Behav 64: 283–290. Swenson JE, Gertsl N, Dahle B & Zedrosser A (2000) Action plan for the conservation of the brown bear (Ursus arctos) in Europe. Strasbourg, Council of Europe Press. 57 Teerink BJ (1991) Hair of West-European mammals. Cambridge, Cambridge University Press. Theuerkauf J, Rouys S & Jedrzejewski W (2003) Selection of den, rendezvous and resting sites by wolves in the Bialowieza Forest, Poland. Can J Zool 81: 163–167. Treves A & Karanth KU (2003) Human-carnivore conflict and perspectives on carnivore management worldwide. Conserv Biol 17: 1491–1499. Tyers DB (2003) Winter ecology of moose on the northern Yellowstone winter range. PhD thesis. Montana State University. Vander Wall SB (1990) Food hoarding in animals. Chicago, University of Chicago Press. van Dijk J, Andersen T, May R, Andersen R & Landa A (2008a) Foraging strategies of wolverines within a predator guild. Can J Zool 86: 966–975. van Dijk J, Gustavsen L, Mysterud A, May R, Flagstad Ø, Broseth H, Andersen R, Andersen R, Steen H & Landa A (2008b) Diet shift of a facultative scavenger, the wolverine, following recolonization of wolves. J Anim Ecol 77: 1183–1190. van Dijk J, Hauge K, Landa A, Andersen R & May R (2007) Evaluating scat analysis methods to assess wolverine (Gulo gulo) diet. Wildlife Biol 13: 62–67. Vangen KM, Persson J, Landa A, Andersen R & Segerström P (2001) Characteristics of dispersal in wolverines. Can J Zool 79: 1641–1649. van Oosterhout C, Hutchinson WF, Wills DPM & Shipley P (2004) MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Mol Ecol Notes 4: 535–538. Vila C, Sundqvist AK, Flagstad Ø, Seddon J, Björnerfeldt S, Kojola I, Casulli A, Sand H, Wabakken P & Ellegren H (2003) Rescue of a severely bottlenecked wolf (Canis lupus) population by a single immigrant. Proc R Soc Lond B Biol 270: 91–97. Walker CW, Vila C, Landa A, Lindén M & Ellegren H (2001) Genetic variation and population structure in Scandinavian wolverine (Gulo gulo) populations. Mol Ecol 10: 53–63. Waples RS (2006) A bias correction for estimates of effective population size based on linkage disequilibrium at unlinked gene loci. Conserv Genet 7: 167–184. Waples RS & Do C (2008) LDNe: a program for estimating effective population size from data on linkage disequilibrium. Mol Ecol Resour 8: 753–756. Weir BA & Cockerham CC (1984) Estimating F-statistics for the analysis of population structure. Evolution 38: 1358–1370. White K, Golden H, Hundertmark K & Lee G (2002) Predation by wolves, Canis lupus, on wolverines, Gulo gulo, and an American marten, Martes Americana, in Alaska. Can Field Nat 116: 132–134. Whitman JS, Ballard WB & Gardner CL (1986) Home range and habitat use by wolverines in southcentral Alaska. J Wildlife Manage 50(3): 460–463. Wikman M (ed.) (2010) Riistakannat 2010. Riistaseurantojen tulokset. Riista- ja kalatalouden tutkimuslaitos, selvityksiä 21/2010. Helsinki, Riista- ja kalatalouden tutkimuslaitos. (In Finnish) 58 Wilmers CC & Post E (2006) Predicting the influence of wolf-provided carrion on scavenger community dynamics under climate change scenarios. Global Change Biol 12: 403–409. Wilmers CC, Stahler DR, Crabtree RL, Smith DW & Getz WM (2003) Resource dispersion and consumer dominance: scavenging at wolf- and hunter-killed carcasses in Greater Yellowstone, USA. Ecol Lett 6: 996–1003. Wilson GA & Rannala B (2003) Bayesian inference of recent migration rates using multilocus genotypes. Genetics 163: 1177–1191. Woodroffe R (2000) Predators and people: using human densities to interpret declines of large carnivores. Anim Conserv 3: 165–173. Woodroffe R & Ginsberg JR (1998) Edge effects and the extinction of populations inside protected areas. Science 280: 2126–2128. Wright JD & Ernst J (2004) Wolverine, Gulo gulo luscus, resting sites and caching behavior in the boreal forest. Can Field Nat 118(1): 61–64. Wright PL & Rausch R (1955) Reproduction of wolverine, Gulo gulo. J Mammal 36: 346– 355. 59 60 Original articles I Koskela A, Kaartinen S, Aspi J, Kojola I, Helle P & Rytkönen S (2013) Does grey wolf presence affect the habitat selection on wolverines? Annales Zoologici Fennici 50: 216–224. II Koskela A, Kojola I, Aspi J & Hyvärinen M (2013) The diet of breeding female wolverines (Gulo gulo) in two areas of Finland. Acta Theriologica 58: 199–204. III Koskela A, Kojola I, Aspi J & Hyvärinen M (2013) Effect of reproductive status on the diet composition of wolverines (Gulo gulo) in boreal forests of eastern Finland. Annales Zoologici Fennici 50: 100–106. IV Esparza-Salas R, Joensuu M, Koskela A, Ollila T & Aspi J Population genetic structure and genetic diversity of the wolverine (Gulo gulo) in Finland. (Manuscript) Reprinted with permission from Finnish Zoological and Botanical Publishing Board (I, III) and Springer (II). Original publications are not included in the electronic version of the dissertation. 61 62 ACTA UNIVERSITATIS OULUENSIS SERIES A SCIENTIAE RERUM NATURALIUM 599. Leiviskä, Katja (2012) Why information systems and software engineering students enter and leave their study programme : A factor model and process theory 600. Siira, Tuula (2012) Value Creation by Enterprise Systems Value Added Resellers : The Case of PLM Systems VARs 601. Kontula, Jukka (2012) New venture creation in software business : A contextually embedded entrepreneur’s perspective 602. Juntunen, Kaisu (2012) Tieto- ja viestintätekniikan soveltamiseen perustuvat toimintaprosessien uudistukset terveydenhuollossa : Sosio-teknis-taloudellinen näkökulma 603. Seppä, Karri (2012) Quantifying regional variation in the survival of cancer patients 604. Kuvaja, Pasi (2012) Software process capability and maturity determination : BOOTSTRAP methodology and its evolution 605. Laanti, Maarit (2012) Agile Methods in large-scale software development organizations : applicability and model for adoption 606. Hokkanen, Juho (2013) Liquid chromatography/mass spectrometry of bioactive secondary metabolites – in vivo and in vitro studies 607. Kuokkanen, Matti (2013) Development of an eco- and material-efficient pellet production chain—a chemical study 608. Jansson, Eeva (2013) Past and present genetic diversity and structure of the Finnish wolf population 609. Myllykoski, Matti (2013) Structure and function of the myelin enzyme 2′,3′-cyclic nucleotide 3′-phosphodiesterase 610. Lehto, Tuomas (2013) The importance of persuasive systems design in enhancing consumers’ perceptions and adoption of health behavior change support systems 611. Hernoux-Villière, Audrey (2013) Catalytic depolymerisation of starch-based industrial waste : use of non-conventional activation methods and novel reaction media 612. Lawrence, Carl (2013) Innovating with information technology in a globalized world : being proactive about culture 613. Ardanov, Pavlo (2013) Priming capacities of endophytic Methylobacterium sp. on potato (Solanum tuberosum L.) Book orders: Granum: Virtual book store http://granum.uta.fi/granum/ A 614 OULU 2013 UNIV ER S IT Y OF OULU P. O. BR[ 00 FI-90014 UNIVERSITY OF OULU FINLAND U N I V E R S I TAT I S S E R I E S SCIENTIAE RERUM NATURALIUM Professor Esa Hohtola HUMANIORA University Lecturer Santeri Palviainen TECHNICA ACTA WOLVERINE HABITAT SELECTION, DIET AND CONSERVATION GENETICS Postgraduate research fellow Sanna Taskila MEDICA Professor Olli Vuolteenaho SCIENTIAE RERUM SOCIALIUM University Lecturer Hannu Heikkinen SCRIPTA ACADEMICA Director Sinikka Eskelinen OECONOMICA Professor Jari Juga EDITOR IN CHIEF Professor Olli Vuolteenaho PUBLICATIONS EDITOR Publications Editor Kirsti Nurkkala ISBN 978-952-62-0226-6 (Paperback) ISBN 978-952-62-0227-3 (PDF) ISSN 0355-3191 (Print) ISSN 1796-220X (Online) UN NIIVVEERRSSIITTAT ATIISS O OU ULLU UEEN NSSIISS U Anni Koskela E D I T O R S Anni Koskela A B C D E F G O U L U E N S I S ACTA A C TA A 614 UNIVERSITY OF OULU GRADUATE SCHOOL; UNIVERSITY OF OULU, FACULTY OF SCIENCE, DEPARTMENT OF BIOLOGY A SCIENTIAE RERUM RERUM SCIENTIAE NATURALIUM NATURALIUM